Embed Size (px)
Citation preview
Guile-CVEdition 0.2.1, revision 1, for use with Guile-CV 0.2.1
The Guile-CV Developpers
This manual documents Guile-CV version 0.2.1.
Copyright (C) 2016 - 2018 Free Software Foundation, Inc.
Permission is granted to copy, distribute and/or modify this document under the terms ofthe GNU Free Documentation License, Version 1.3 or any later version published by the FreeSoftware Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-CoverTexts. A copy of the license is included in the section entitled “GNU Free DocumentationLicense.”
i
Table of Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Contributors to this Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1The Guile-CV License . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 About Guile-CV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Obtaining and Installing Guile-CV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.3 Contact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.4 Reporting Bugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Using Guile-CV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.1 Configuring Guile for Guile-CV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.1 Configuring Guile’s repl-print procedure . . . . . . . . . . . . . . . . . . . . 62.1.2 Configuring Guile’s raised exception system . . . . . . . . . . . . . . . . 6
2.2 Images used in Guile-CV’s documentation . . . . . . . . . . . . . . . . . . . . . . . 82.3 Starting Guile-CV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3 API Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93.1 API Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.1.1 Naming Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93.1.2 Abreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.2 Guile-CV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103.2.1 Image Structure and Accessors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103.2.2 Kernel Structure and Accessors . . . . . . . . . . . . . . . . . . . . . . . . . . . 123.2.3 Import Export . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183.2.4 Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.2.5 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263.2.6 Texture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293.2.7 Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303.2.8 Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333.2.9 Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363.2.10 Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.3 Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413.3.1 Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413.3.2 Goops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413.3.3 G-export . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413.3.4 Pi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413.3.5 Utils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Appendix A GNU Free Documentation License . . 42
ii
Concept Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Procedure Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Variable Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Type Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
1
Preface
This manual describes how to use Guile-CV. It relates particularly to Guile-CV version0.2.1.
Contributors to this Manual
Like Guile-CV itself, the Guile-CV reference manual is a living entity. Right now, thecontributor to this manual is:
• David Pirotte
who is also the author and maintainer of Guile-CV.
You are most welcome to join and help. Visit Guile-CV (http://www.gnu.org/software/guile-cv)web site to find out how to get involved.
The Guile-CV License
Guile-CV is Free Software. Guile-CV is copyrighted, not public domain, and there arerestrictions on its distribution or redistribution:
• Guile-CV and supporting files are published under the terms of the GNU GeneralPublic License version 3 or later. See the file LICENSE.
• This manual is published under the terms of the GNU Free Documentation License(see Appendix A [GNU Free Documentation License], page 42).
You must be aware there is no warranty whatsoever for Guile-CV. This is described in fullin the license.
1 Introduction
1.1 About Guile-CV
GNU Guile-CVImage Processing and Analysis in Guile (http://www.gnu.org/software/guile)a Computer Vision functional programming library
Guile-CV (http://www.gnu.org/software/guile-cv) - Image Processing and Analysis inGuile (http://www.gnu.org/software/guile) - is a Computer Vision functional program-ming library for the Guile (http://www.gnu.org/software/guile) Scheme language.
Guile-CV (http://www.gnu.org/software/guile-cv) is based on Vigra(http://ukoethe.github.io/vigra/) (Vision with Generic Algorithms), en-hanced with additional algorithms (Image Textures, Delineate, Reconstruction andmany more), all accessible through a nice, clean and easy to use high level API.
Guile-CV (http://www.gnu.org/software/guile-cv) is natively multi-threaded, andtakes advantage of multiple cores, using high-level and fine grained application-levelparallelism constructs available in Guile (http://www.gnu.org/software/guile),based on its support to POSIX threads.
Chapter 1: Introduction 2
Objective
Guile-CV (http://www.gnu.org/software/guile-cv) objective is to be a robust,reliable and fast - Image Processing and Analysis - Computer Vision functionalprogramming library for the Guile (http://www.gnu.org/software/guile) Schemelanguage. Guile-CV (http://www.gnu.org/software/guile-cv) also wants to be easy
to use, study, modify and extend.
Guile-CV (http://www.gnu.org/software/guile-cv) can be used as an educational
sofware, a research toolbox but it can also be used ’in production’: Guile-CV(http://www.gnu.org/software/guile-cv) is robust, reliable and fast, and wewill make sure Guile-CV (http://www.gnu.org/software/guile-cv) remains robust,reliable and fast as it grows.
Savannah
Guile-CV (http://www.gnu.org/software/guile-cv) also has a Savannah project page(http://savannah.gnu.org/projects/guile-cv).
1.2 Obtaining and Installing Guile-CV
Guile-CV (http://www.gnu.org/software/guile-cv) can be obtained from the follow-ing archive site (http://ftp.gnu.org/gnu/guile-cv). The file will be named guile-cv-version.tar.gz. The current version is 0.2.1, so the file you should grab is:
http://ftp.gnu.org/gnu/guile-cv/guile-cv-0.2.1.tar.gz
Dependencies
Guile-CV (http://www.gnu.org/software/guile-cv) needs the following software to run:
• Autoconf >= 2.69
• Automake >= 1.14
• Makeinfo >= 6.3
• Guile (http://www.gnu.org/software/guile) >= 2.0.13
• Guile-Lib (http://www.nongnu.org/guile-lib) >= 0.2.5
• Vigra (http://ukoethe.github.io/vigra/) >= 1.11.0
Note:
If you manually install Vigra (http://ukoethe.github.io/vigra/), make sureyou pass the cmake DCMAKE BUILD TYPE=RELEASE option, which triggersabsolutely essential adequate runtime optimization flags.
• Vigra C (https://github.com/BSeppke/vigra_c) >= commit 0af647d08 - Oct 28,2018
The local minima and maxima interfaces have been improved, and now supportthe full set of options provided by Vigra, to our request (thank you Benjamin!).In addition there has been a few bugs fixed, including one we detected whileworking on Guile-CV local minima bindigs.
Vigra C - a C wrapper [to some of] the Vigra functionality - is currently only avail-able by cloning its source code git (https://git-scm.com/) repository: there is
Chapter 1: Introduction 3
no release and no versioning scheme either1. But no big deal, its home page hasan ’Installation’ section which guides you step by step.
Notes:
1. Make sure you pass the cmake DCMAKE BUILD TYPE=RELEASE op-tion, which triggers absolutely essential adequate runtime optimization flags;
2. Vigra C says it depends on cmake >= 3.1, but this is only true if you want tobuild its documentation, probably not the case. Most distribution still havecmake 2.8, if that is your case, you may safely edit /your/path/vigra_
c/CMakeLists.txt and downgrade this requirement to the cmake versioninstalled on your machine;
3. Make sure the directory where libvigra_c.so has been installed is’known’, either because it is defined in /etc/ld.so.conf.d, or you set theenvironment variable LD_LIBRARY_PATH, otherwise Guile won’t find it andconfigure will report an error.
• LaTex (http://www.latex-project.org/)
Any modern latex distribution will do, we use TexLive (https://tug.org/texlive/).
Guile-CV will check that it can find the standalone documentclass, as well as thefollowing packages: inputenc, fontenc, lmodern, xcolor, booktabs, siunitx,iwona.
Iwona (http://www.tug.dk/FontCatalogue/iwona/): this is the font used tocreate [im-histogram], page 39, headers, legend indices and footers. Note that itcould be that it is not part of your ’basic’ LaTex distro, on debian for example,iwona is part of the texlive-fonts-extra package.
Install from the tarball
Assuming you have satisfied the dependencies, open a terminal and proceed with the fol-lowing steps:
cd <download-path>
tar zxf guile-cv-0.2.1.tar.gz
cd guile-cv-0.2.1
./configure [--prefix=/your/prefix] [--with-guile-site=yes]
make
make install
Special note:
Before you start to use Guile-CV (http://www.gnu.org/software/guile-cv), makesure you read and implement the recommendation made in the manual, section SeeSection 2.1 [Configuring Guile for Guile-CV], page 6.
Happy Guile-CV (http://www.gnu.org/software/guile-cv)!
1 We do our best to check that the libvigra c installed library does contain the required Guile-CV functionaltythough, and these checks are listed as part of our configure steps
Chapter 1: Introduction 4
Install from the source
Guile-CV (http://www.gnu.org/software/guile-cv) uses Git (https://git-scm.com/)for revision control, hosted on Savannah (http://savannah.gnu.org/projects/guile-cv),you may browse the sources repository here (http://git.savannah.gnu.org/cgit/guile-cv.git).
There are currently 2 [important] branches: master and devel. Guile-CV(http://www.gnu.org/software/guile-cv) stable branch is master, developments occuron the devel branch.
So, to grab, compile and install from the source, open a terminal and:
git clone git://git.savannah.gnu.org/guile-cv.git
cd guile-cv
./autogen.sh
./configure [--prefix=/your/prefix] [--with-guile-site=yes]
make
make install
Special note:
Before you start to use Guile-CV (http://www.gnu.org/software/guile-cv), makesure you read and implement the recommendation made in the manual, section SeeSection 2.1 [Configuring Guile for Guile-CV], page 6.
The above steps ensure you’re using Guile-CV (http://www.gnu.org/software/guile-cv)bleeding edge stable version. If you wish to participate to developments, checkout thedevel branch:
git checkout devel
Happy hacking!
Notes:
1. The default and --prefix installation locations for source modules and compiled files(in the absence of --with-guile-site=yes) are:
$(datadir)/guile-cv
$(libdir)/guile-cv/guile/$(GUILE_EFFECTIVE_VERSION)/site-ccache
If you pass --with-guile-site=yes, these locations become the Guile global site andsite-ccache directories, respectively.
The configure step reports these locations as the content of the sitedir andsiteccachedir variables, respectivelly the source modules and compiled files installlocations. After installation, you may consult these variables using pkg-config:
pkg-config guile-cv-1.0 --variable=sitedir
pkg-config guile-cv-1.0 --variable=siteccachedir
You will need - unless you have used --with-guile-site=yes, orunless these locations are already ’known’ by Guile - to define oraugment your GUILE_LOAD_PATH and GUILE_COMPILED_PATH environ-ment variables with these locations, respectively (or %load-path and%load-compiled-path at run time if you prefer2 (See Environment Variables
2 In this case, you may as well decide to either alter your $HOME/.guile personal file, or, if you are workingin a mult-user environmet, you may also opt for a global configuration. In this case, the file must be
Chapter 1: Introduction 5
(https://www.gnu.org/software/guile/manual/guile.html#Environment-Variables)and Load Path (https://www.gnu.org/software/guile/manual/guile.html#Load-Paths)in the Guile Reference Manual).
2. Guile-CV also installs its libguile-cv.* library files, in $(libdir). The configurestep reports its location as the content of the libdir variable, which depends on onthe content of the prefix and exec_prefix variables (also reported). After nstallation,you may consult these variables using pkg-config:
pkg-config guile-cv-1.0 --variable=prefix
pkg-config guile-cv-1.0 --variable=exec_prefix
pkg-config guile-cv-1.0 --variable=libdir
You will need - unless the $(libdir) location is already ’known’ by your system - toeither define or augment your $LD_LIBRARY_PATH environment variable, or alter the/etc/ld.so.conf (or add a file in /etc/ld.so.conf.d) and run (as root) ldconfig,so that Guile-CV finds its libguile-cv.* library files3.
3. To install Guile-CV, you must have write permissions to the default or $(prefix)
directory and its subdirs, as well as to both Guile’s site and site-ccache directories if--with-guile-site=yes was passed.
4. Like for any other GNU Tool Chain compatible software, you may install thedocumentation locally using make install-info, make install-html and/or make
install-pdf.
5. Last but not least :), Guile-CV comes with a test-suite, which we recommend youto run (especially before Section 1.4 [Reporting Bugs], page 6):
make check
1.3 Contact
Mailing lists
Guile-CV uses the following mailing list:
• [email protected] is for general user help and discussion.
• [email protected] is used to discuss most aspects of Guile-CV, including develop-ment and enhancement requests.
When sending emails to guile-user and/or guile-devel, please use ’Guile-CV: ’ to prefix thesubject line of any Guile-CV related email, thanks!
IRC
Most of the time you can find me on irc, channel #guile, #guix and #scheme onirc.freenode.net, #clutter and #introspection on irc.gnome.org, under the nicknamedaviid.
named init.scm and placed it here (evaluate the following expression in a terminal): guile -c "(display
(%global-site-dir))(newline)".3 Contact your administrator if you opt for the second solution but don’t have write priviledges on your
system.
Chapter 2: Using Guile-CV 6
1.4 Reporting Bugs
Guile-CV uses the following bug reports mailing list:
You can (un)subscribe to the bugs report list by following instructions on the list informationpage (https://lists.gnu.org/mailman/listinfo/bug-guile-cv).
2 Using Guile-CV
Guile-CV Reference Manual still is a mock-up: any help is more then welcome to improvethis situation, thanks!
2.1 Configuring Guile for Guile-CV
Guile must be modified, with respect to two core functionalities, before to start to useGuile-CV: (a) its repl-print procedure and (b) its raised exception system.
2.1.1 Configuring Guile’s repl-print procedure
Guile’s repl-print procedure calls (write val), which is inadequate for images, even verysmall images4. Fortunately, Guile provides us both a simple mechanism to alter the defaultrepl printer and the alternate repl printer procedure we need: truncated-print.
To modify the default repl printer, you have to modify your $HOME/.guile or, if you areworking in a multi-user environmet, you may choose to add (or modify if it exists) the filenamed init.scm in the so-called Guile global site directory5, here (evaluate the followingexpression in a terminal): guile -c "(display (%global-site-dir))(newline)".
Which ever solution you choose, add the following lines:
(use-modules (ice-9 pretty-print)
(system repl common))
(repl-default-option-set! ’print
(lambda (repl obj)
(truncated-print obj) (newline)))
2.1.2 Configuring Guile’s raised exception system
Guile’s default raised exception system calls simple-format, which is inadequate for images,even very small images (see the related footnote of the previous section, it explains how‘inadequate’ this default is for images.
Because Guile’s module we need to patch has changed in between 2.0, 2.2 and 2.2.3, we willguide you to manually update your local version, just don’t be scared, be meticulous and it
4 Even for very small images, using write is inadequate, in a terminal, and will definitely kill your Emacs/Geisersession. Not to mention it will raise your electricity bill :) - till you succeed to delete its process, Emacswill use one core at more then 100%, desperately trying to display hundreds of thousands of floating pointvalues, heating your laptop (if you have a laptop) up to the point you’ll be able to cook an egg on it, andget its fans crasy... You’ve been warned :).
5 You need write privileges to add or modify this file, contact your system administrator if you’re not in chargeof the system you are working on.
Chapter 2: Using Guile-CV 7
will be all fine. But if you think it is ’too much’ for you, get in touch with us, and we willguide you or provide a ’ready to use module’, depending on your version of Guile.
The module we need to modify is (ice-9 boot-9), so let’s first figure out where it is onyour system6, in a guile session, enter the following:
(string-append (%package-data-dir) "/" (effective-version))
a$2 = "/opt2/share/guile/2.2"
The above returned value is an example of course, just proceed with the value returned byyour system. So, the file we need to edit, in our example, is here:
/opt2/share/guile/2.2/ice-9/boot-9.scm
Edit the above file and:
1. Search for the line (define format simple-format) (for Guile 2.2.3, it is the line 327),and below, add the following lines:
;; instead of using the above, let’s define a specific format binding
;; for exception printers, to allow its user customization.
(define exception-format simple-format)
2. In the core of the following procedure, (and only in the core of the following procedures),you will replace all occurences calls to format by calls to exception-format:
dispatch-exception - line 731 for Guile 2.2.3
1 occurrence
(let ((exception-printers ’())) - line 864 for Guile 2.2.3
5 occurrences
scm-error-printer - line 910 for Guile 2.2.3
2 occurences
syntax-error-printer - line 921 for Guile 2.2.3
7 occurences
keyword-error-printer - line 941 for Guile 2.2.3
1 occurrence
getaddrinfo-error-printer - line 941 for Guile 2.2.3
1 occurrence
false-if-exception - line 1063 for Guile 2.2.3
2 occurences
make-record-type - line 1200 for Guile 2.2.3
1 occurence
6 You need write privileges to modify this module, contact your admin if you’re not in charge of the systemyou are working on.
Chapter 2: Using Guile-CV 8
3. Let’s compile it, and install the compiled version where it should (in the following lines,you’ll need to substitute /opt2 by your $prefix value, and the username by yours):
cd /opt2/share/guile/2.2/ice-9
guild compile boot-9.scm
a ;;; note: source file /opt2/share/guile/2.2/ice-9/boot-9.scm
;;; newer than compiled /opt2/lib/guile/2.2/ccache/ice-9/boot-9.go
wrote ‘$HOME/.cache/guile/ccache/2.2-LE-8-3.A/opt2/share/guile/2.2/ice-9/boot-9.scm.go’
Now we move it, so all users benefit from the new compiled version:
mv $HOME/.cache/guile/ccache/2.2-LE-8-3.A/opt2/share/guile/2.2/ice-9/boot-9.scm.go /opt2/lib/guile/2.2/ccache/ice-9/boot-9.go
Now, once the above is completed, you need to add the following lines (see below) to your$HOME/.guile or, if you are working in a multi-user environmet, to the file named init.scm
in the so-called Guile global site directory (the previous subsection lists the terminal com-mand you need to run to see where that directory is on your system).
Note that the following proposed modification assumes you did perform the steps rec-ommended in the previous subsection, which means that (use-modules (ice-9 pretty-
print)) is already in there:
(when (defined? ’exception-format)
(set! exception-format
(lambda (port fmt . args)
(for-each (lambda (arg)
(truncated-print arg #:port port))
args))))
You are now ready to use Guile-CV!
2.2 Images used in Guile-CV’s documentation
All images used in Guile-CV’s documentation are distributed with the source and installedhere:
$prefix/share/doc/guile-cv/images
All examples using im-load and im-save given in this manual, unless a full pathname isspecified, assume that these images are available from the guile current working directory,see getcwd and chdir in Guile’s manual
Our best recommendation, at least to start with, is to create a working directory, such asmkdir $HOME/guile-cv/images, for example, and as you need them, copy the distributedimages you are interested in.
2.3 Starting Guile-CV
Special note:
Before you start to use Guile-CV, make sure you read and implement the recommen-dation made in Section 2.1 [Configuring Guile for Guile-CV], page 6,
With the previous Section 2.2 [Images used in Guile-CV’s documentation], page 8, recom-mendations in mind, open a terminal and:
cd ~/guile-cv/images
Chapter 3: API Reference 9
guile
scheme@(guile-user)> ,use (cv)
scheme@(guile-user)> (im-load "sand.tif")
a$2 = (512 512 1 (#f32(125.0 128.0 124.0 118.0 108.0 75.0 76.0 # ...)))
Or if you use Emacs (https://www.gnu.org/software/emacs) which, coupled with Geiser(http://www.nongnu.org/geiser) absolutely rocks :-), then a typical session becomes:
fire Emacs
M-x cd
aChange default directory: ~/guile-cv/images
M-x run-guile
scheme@(guile-user)> ,use (cv)
scheme@(guile-user)> (im-load "sand.tif")
a$2 = (512 512 1 (#f32(125.0 128.0 124.0 118.0 108.0 75.0 76.0 # ...)))
Note that to benefit from Emacs’s Tab completion mechanism, while typing image filenames,Emacs itself must be in that directory, hence the above first step M-x cd ...
3 API Reference
3.1 API Overview
Guile-CV Low Level API
The Guile-CV low level API is ...
Guile-CV High Level API
Guile-CV will also provide a higher level API ...
3.1.1 Naming Conventions
Vigra Funtions
Guile-CV low level API procedure names hat bind a Vigra function always start with vigra-
...
bluefox
->
redbear
3.1.2 Abreviations
In the table below we list respectively the abreviation ...
Chapter 3: API Reference 10
3.2 Guile-CV
3.2.1 Image Structure and Accessors
The Guile-CV procedures and methods related to images data structure, creating, accessingand copying images.
Image Data Structure
A Guile-CV image is represented by a list containing the following elements:
(width height n-channel idata)
where idata is a list of n-channel elements, each element being a vector of (* width height)
cells. More precisely, each element is an srfi-4 homogeneous numeric vector of 32 bit floats,called f32vector, knowing that f32 is the C type float.
The external representation (ie. read syntax) for idata vectors is #f32(...). As an example,a gray scale image of width 3 and height 2, initialized to 0.0 is represented by the followingexpression:
(3 2 1 (#f32(0.0 0.0 0.0 0.0 0.0 0.0)))
The n-channel is an integer >= 1, with no limit but the memory size. This said, mostGuile-CV procedures and methods expect either GRAY scale (n-channel=1), or RGB (n-channel=3) images. For the later, the channels are Red, Green and Blue in that order.
Guile-CV provides usefull accessors for all these fields. However, very often, you will needthem all, in which case your best friend is (ice-9 match), here is an example:
,use (cv)
(define image (im-make 4 3 3))
(match image
((width height n-chan idata)
(match idata
((r g b)
... your code here ...))))
You will find many examples of such a ‘pattern’ in Guile-CV’s source code itself of course,along with some other ‘techniques’ that might be useful, so we invite you to read it, and ifyou do so: feedback, design and code review is more then welcome! This section describeswhat is in the module (cv idata).
Note that the (cv) module imports and re-exports, among may others, the public interfaceof (ice-9 match).
Procedures
[Procedure]im-make width height n [value][Procedure]im-make-channel width height [value][Procedure]im-make-channels width height n [value]
Returns a new image, list of channels or channel.
Each channel is an srfi-4 homogeneous vector of 32 bit floats (f32vector), of width byheight initialized to value. The default value is 0.0
Chapter 3: API Reference 11
[Procedure]im-copy image[Procedure]im-copy-channel channel width height
Returns a new fresh copy of image or channel.
[Method]im-size imageReturns the list of (width height n-channel)for image.
[Method]im-width image[Method]im-height image[Method]im-n-channel image[Method]im-channels image
[Procedure]im-channel image nReturns, respectively the width, the height, n-channel, channels or the nth channelfor image.
[Procedure]im-image? image[Method]im-gray? image[Method]im-rgb? image
Returns #t if image is respectively a Guile-CV image, a GRAY scale or an RGBimage.
[Procedure]im-binary? i1 i2 i3 . . .[Procedure]im-binary-channel? width height c1 c2 c3 . . .
Returns #t if i1 i2 i3 . . . or c1 c2 c3 . . . respectively are BINARY (Black and White)images or channels respectively.
Note that when more then one image or channel is passed, they must all be of thesame size.
[Procedure]im-=? [precision] i1 i2 i3 . . .[Procedure]im-=-channel? width height [precision] c1 c2 c3 . . .
Returns #t if i1 i2 i3 . . . or c1 c2 c3 . . . respectively are of the same size, have thesame number of channels that all respectively contain the same values.
If the first argument is a number, it is used as the precision to compare pixel values.The default precision value is 1.0e-4. Note that if you are certain your images orchannels contain ’discrete’ float values, you may pass 0.0 as the precision to be used,i which case values will be compared using = (instead of float=?, which is faster.
[Procedure]im-ref image i j [k][Procedure]im-fast-ref image i j [k]
Returns the pixel value stored at position i and j of the image channel k. The defaultvalue for k is 0.
im-fast-ref does not check the validity of its arguments: use it at your own risk.
[Procedure]im-set! image i j [k] value[Procedure]im-fast-set! image i j [k] value
Returns nothing.
Sets the pixel value stored at position i and j of the image channel k to value. Thedefault value for k is 0.
im-fast-set! does not check the validity of its arguments: use it at your own risk.
Chapter 3: API Reference 12
[Procedure]im-channel-offset i j width height[Procedure]im-fast-channel-offset i j width
Returns the channel offset for the i and j indices, based on the width and height ofthe channel.
This procedure converts the matrix indices i and j to a vector offset for a channel ofsize width and height.
im-fast-channel-offset does not check the validity of its arguments: use it at your ownrisk.
[Procedure]im-channel-ref channel i j width height[Procedure]im-fast-channel-ref channel i j width
Returns the pixel value stored at position i and j of the channel of size width andheight.
im-fast-channel-ref does not check the validity of its arguments: use it at your ownrisk.
[Procedure]im-channel-set! channel i j width height value[Procedure]im-fast-channel-set! channel i j width value
Returns nothing.
Sets the pixel at position i and j of channel of size width and height to value.
im-fast-channel-set! does not check the validity of its arguments: use it at your ownrisk.
[Procedure]im-collect what i1 i2 i3 . . .Returns a list of what collected from i1 i2 i3 . . .
The valid what synbols are:
size
width
height
n-channel
channels
chan-0, gray, red
chan-1, green
chan-2, blue
chan-k (*)
(*): whith k being a valid channel indice, [0 (- n 1)].
3.2.2 Kernel Structure and Accessors
The Guile-CV procedures and methods related to kernels data structure, creating andaccessing kernels.
Chapter 3: API Reference 13
Kernel Data Structure
A Guile-CV kernel (https://en.wikipedia.org/wiki/Kernel_(image_processing)) isrepresented by a list containing the following elements:
(width height kdata)
where kdata is a vector of (* width height) cells. More precisely, kdata is an srfi-4
homogeneous numeric vector of 64 bit floats, called f64vector, knowing that f64 is the Ctype double.
The external representation (ie. read syntax) for kdata vectors is #f64(...). As an exam-ple, the identity kernel of width 3 and height 3, initialized to 0.0 is represented by thefollowing expression:
(3 3 #f64(0.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0))
The kernel width and height can be different (kernels can be rectangular), but both widthand height must be odd values.
Guile-CV provides useful accessors for kernel fields, however, if you need them all, just likefor accessing image fields, your best friend is (ice-9 match), here is an example:
,use (cv)
(match kernel
((width height kdata)
... your code here ...))
Note that the (cv) module imports and re-exports, among may others, the public interfaceof (ice-9 match).
Guile-CV defines a few useful kernels, see [kernel variables], page 16, at the end of thissection, that you both may want to use and reuse: it will be easier, if you need to do so, todefine your own kernels reusing an existing one, see the (cv kdata) module.
Procedures
[Procedure]k-make width height [values #f] [norm #f]Returns a new kernel.
The kdata value of this new kernel is an srfi-4 homogeneous numeric vector of 64 bitfloats, f64vector, composed of width by height cells.
The optional values argument can be:
#f kdata is initialized to the ‘identity’ kernel (all zeros except thecenter of the kernel, initialzed to 1)
a single value
all kdata cells are initialized using that single value
a list of values
a list of width by height values, used to initialzed kdata, in theorder they are given
The optional norm argument can be:
#f in this case, kdata is not normalized
Chapter 3: API Reference 14
#t unless values would be #f, kdata is normalized using (reduce + 0
values)
a single value
all kdata cells are normalized using that value, which must be anumber different from 0
When both values and norm are passed - which is mandatory if you want to passnorm (since these are optional arguments, as opposed to keyword arguments) - valuesmust precede norm on the arguments list.
As an example, here is how to define a 3 x 3 normalized mean kernel:
,use (cv)
(k-make 3 3 1 #t)
a$2 = (3 3 #f64(0.1111111111111111 0.1111111111111111 # # # # ...))
(k-display $2)
a
0.11111 0.11111 0.11111
0.11111 0.11111 0.11111
0.11111 0.11111 0.11111
[Procedure]k-make-circular-mask radius [value 1] [norm #f]Returns a new circular mask kernel.
The kdata value of this new kernel is an srfi-4 homogeneous numeric vector of 64bit floats, f64vector, composed of width by height cells where width and height areequal and odd values determined by the procedure.
The mandatory radius argument must be a floating point number satisfying thefollowing predicate: (float>=? radius 0.5).
The optional norm argument can be:
#f in this case, kdata is not normalized
#t kdata values are normalized using (* n value), where n is the num-ber of non zero elements of the circular kernel mask being defined.
When both value and norm are passed - which is mandatory if you want to pass norm(since these are optional arguments, as opposed to optional keyword arguments) -value must precede norm on the arguments list.
To illustrate, here are the circular kernel masks of radius 0.5, 1, 1.5 respectively:
...
(for-each (lambda (i)
(k-display (k-make-circular-mask i)
#:proc float->int))
’(0.5 1.0 1.5))
a
0 1 0
Chapter 3: API Reference 15
1 1 1
0 1 0
1 1 1
1 1 1
1 1 1
0 0 1 0 0
0 1 1 1 0
1 1 1 1 1
0 1 1 1 0
0 0 1 0 0
To better illustrate, let’s define a bigger circular kernel mask, transform it to an imageand [im-show], page 40, it:
...
(match (k-make-circular-mask 49)
((w h kdata) (list w h 1 (list (f64vector->f32vector kdata)))))
a$6 = (99 99 1 (#f32(0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 # ...)))
(im-show $6 ’scale)
And you should see the following image7
[Procedure]k-width kernel[Procedure]k-height kernel[Procedure]k-size kernel[Procedure]k-channel kernel
Returns, respectively, the width, the height, the list of width and height or the kdatafor kernel.
[Procedure]kernel? kernelReturns #t if kernel is a Guile-CV kernel.
[Procedure]k-ref kernel i j[Procedure]k-fast-ref kernel i j
Returns the value stored at position i and j of the kernel.
k-fast-ref does not check the validity of its arguments: use it at your own risk.
[Procedure]k-set! kernel i j value[Procedure]k-fast-set! kernel i j value
Returns nothing.
Sets the value stored at position i and j of the kernel to value.
k-fast-set! does not check the validity of its arguments: use it at your own risk.
7 The ’scale optional argument passed to [im-show], page 40, as its name indicate, is so that kernel valueswill be scaled, which in this case means that 1.0 values will become 255.0 - otherwise, it would be almostimpossible for a human eye to actually see the shape of the circle . . .
Chapter 3: API Reference 16
[Procedure]k-offset i j width height[Procedure]k-fast-offset i j width
Returns the kernel offset for the i and j indices, based on the width and height of thekernel.
This procedure converts the matrix indices i and j to a vector offset for a kernel ofsize width and height.
k-fast-offset does not check the validity of its arguments: use it at your own risk.
[Procedure]k-display image [#:proc #f] [#:port (current-output-port)]Returns nothing.
Displays the content of kernel on port, applying proc to each kernel value.
,use (cv)
(k-display %k-laplacian)
a0.37500 0.25000 0.37500
0.25000 -2.50000 0.25000
0.37500 0.25000 0.37500
Variables
Notes: (a) the following kernels are merely offered as ‘didactic’ examples, some of thesewere used ‘in the old days’, but in most cases, you will find and prefer to use a ‘specific’and ‘modern’ procedure that will give (much) better results, such as, [im-gaussian-blur],page 30, [im-gaussian-sharp], page 31, [im-sharpen], page 31, (a simple sharpening proce-dure), [im-canny], page 37, (edge detection) ... and (b) in order to make these definitionseasier to read, we’ve added some spaces and newlines.
[Variable]%k-identity(k-display %k-identity #:proc float->int)
a0 0 0
0 1 0
0 0 0
[Variable]%k-edge0(k-make 3 3
’( 1 0 -1
0 0 0
-1 0 1 ))
[Variable]%k-edge1(k-make 3 3
’( 0 1 0
1 -4 1
0 1 0 ))
[Variable]%k-sharpen(k-make 3 3
Chapter 3: API Reference 17
’( -1 -1 -1
-1 9 -1
-1 -1 -1 ))
[Variable]%k-mean(k-make 3 3
’( 1 1 1
1 1 1
1 1 1 )
9)
[Variable]%k-gaussian-blur0(k-make 3 3
’( 1 2 1
2 4 2
1 2 1 )
16)
[Variable]%k-gaussian-blur1(k-make 5 5
’( 1 4 6 4 1
4 16 24 16 4
6 24 36 24 6
4 16 24 16 4
1 4 6 4 1 )
256)
[Variable]%k-unsharp(k-make 5 5
’( 1 4 6 4 1
4 16 24 16 4
6 24 -476 24 6
4 16 24 16 4
1 4 6 4 1 )
-256)
[Variable]%k-embossAlso called %k-compass or %k-directional, this kind of filter is useful to enhanceedges in given directions. With a 3 x 3 kernel, one normally uses filters for 0, 45,90 and 135 degrees. The various angles are obtained ‘rotating’ the positive andnegative values to ‘align’ with the various directions.
(k-make 3 3
’( -2 -2 0
-2 6 0
0 0 0 ))
[Variable]%k-laplacianThis is a variation of the more traditional Laplacian kernels, that are meant to enhanceedges, in this case in an isotropic fashion (non-directional). This the implementation
Chapter 3: API Reference 18
in the Vigra code and it atributes large weights to the diagonal pixels of the kernel.Nevertheless, the total weight is zero.
(k-make 3 3
’( 0.375 0.25 0.375
0.25 -2.5 0.25
0.375 0.25 0.375 ))
Prewitt filtering
[Variable]%k-prewitt-yA 3 x 3 kernel which emphasizes horizontal edges by approximating a vertical gradient.
(k-make 3 3
’( 1 1 1
0 0 0
-1 -1 -1 ))
[Variable]%k-prewitt-xA 3 x 3 kernel which emphasizes vertical edges by approximating an horizontal gra-dient.
(k-make 3 3
’( 1 0 -1
1 0 -1
1 0 -1 ))
Sobel filtering
Filtering an image using a ‘Sobel filter’ requires a three steps approach: (1) filtering theimage using the ‘Sobel y filter’, (2) dito using the ‘Sobel x filter’, and (3) combiningthe results to obtain ‘Sobel magnitude’: (sqrt (+ (sqrt sobel-y) (sqrt sobel-x)).
[Variable]%k-sobel-y(k-make 3 3
’( 1 2 1
0 0 0
-1 -2 -1 ))
[Variable]%k-sobel-x(k-make 3 3
’( 1 0 -1
2 0 -2
1 0 -1 ))
3.2.3 Import Export
The Guile-CV procedures and methods to load, save and query file system images.
Procedures
[Procedure]im-load filenameReturns a Guile-CV image.
Chapter 3: API Reference 19
Loads the image pointed by filename and returns a Guile-CV image. filename caneither be a GRAY or an RGB image.
At this point, Guile-CV supports the following file formats: GIF, TIFF, JPEG, BMP,EXR, HDR, PNM (PBM, PGM, PPM), PNG, SunRaster, KHOROS-VIFF.
[Procedure]im-save image filename [scale #f]Returns #t.
Saves image in filename.
The optional scale argument can take the following values:
#f pixel values are ‘clipped’: values < 0 are saved as 0, values > 255
are saved as 255, and otherwise are saved unchanged
#t all pixel values are scaled8 to the [0 255] range
The type in which image is saved is determined by the filename extension, as in thefolowing example:
(im-load "edx.png")
...
(im-save $4 "/tmp/edx.jpg")
[Method]im-size filenameReturns the list of (width height n-channel)for filename.
[Method]im-width filename[Method]im-height filename[Method]im-n-channel filename
Returns, respectively the width, the height and the n-channel for filename.
[Method]im-gray? filename[Method]im-rgb? filename
Returns #t if filename is respectively a GRAY scale or an RGB image.
3.2.4 Processing
The Guile-CV procedures and methods to process images.
Procedures
[Procedure]im-resize image new-width new-height [#:i-mode ’bilinear][Procedure]im-resize-channel channel width height new-width new-height
[#:i-mode ’bilinear]Returns a new image or chanbnel resized to new-width, new-height.
The interpolation mode #:i-mode, can be one of:
none
bilinear
8 Note that in this particular context, scale does not mean a change in dimension, but rather bringing pixelvalues from the range they occupy in memory to the [0 255] range
Chapter 3: API Reference 20
biquadratic
bicubic
? (fixme)
[Procedure]im-rotate image angle [#:i-mode ’bilinear][Procedure]im-rotate-channel channel width height angle [#:i-mode ’bilinear]
Returns a new image or channel rotated by angle.
The angle is in degrees: +/-[0.0 360.0].
It is neccessary, for rotations other than multiples of 90˚, to recalculate the tar-get coordinates, since after the rotation, they might be floats. The ’next neighbor’interpolation possible modes, #:i-mode, are:
bilinear
biquadratic
bicubic
? (fixme)
[Procedure]im-flip image plane[Procedure]im-flip-channel channel width height plane
Returns a new image or channel flipped according to the selected plane.
Valid flipping plane values are:
hori horizontal
vert vertical
both
[Procedure]im-invert image[Procedure]im-invert-channel channel width height
Returns a new inversed image or channel.
Calculating the inverse of an image or a channel consist of applying the exponentfunction, expt, to all pixel values, raising them to the power of -1.
[Procedure]im-transpose image[Procedure]im-transpose-channel channel width height
Returns a new tranposed image or channel.
Transposing an image or a channel consist of flipping it over its main diagonal. Inthe transposed result, switched in size, row values are the original column values andcolumn values are the original row values.
[Procedure]im-normalize image [#:value 255.0][Procedure]im-normalize-channel channel width height [#:value 255.0]
Returns a new normalized image or channel.
Normalizing an image or a channel consist of dividing all pixels by a value so they allfall in the [0.0 -> 1.0] range. The default #:value is 255.0.
Chapter 3: API Reference 21
[Procedure]im-clip image [#:lower 0.0] [#:upper 255.0][Procedure]im-clip-channel channel width height [#:lower 0.0] [#:upper
255.0]Returns a new clipped image or channel.
Clipping an image or a channel consist of replacing all pixel values below lower bythe lower value and all pixel values above upper by the upper value.
[Procedure]im-crop image left top right bottom[Procedure]im-crop-channel channel width height left top right bottom
[#:new-w #f] [#:new-h #f]Returns a new image, resulting of the crop of image at left, top, right and bottom.
[Procedure]im-crop-size width height left top right bottomReturns a list, (new-width new-height).
Given the original image width and height, this procedure checks that left, top, rightand bottom are valid and return a list, (new-width new-height), otherwise, it raisesan error.
[Procedure]im-padd image left top right bottom [#:color ’(0.0 0.0 0.0)][Procedure]im-padd-channel channel width height left top right bottom
[#:new-w #f] [#:new-h #f] [#:value 0.0]Returns a new image or channel, respectively padding image or channel by left, top,right and bottom pixels initialized respectively to color or value. Note that whenim-padd is called upon a GRAY image, color is reduced to its corresponding grayvalue:
(/ (reduce + 0 color) 3)
[Procedure]im-padd-size width height left top right bottomReturns a list, (new-width new-height).
Given the original image width and height, this procedure checks that left, top, rightand bottom are >= 0 and return a list, (new-width new-height), otherwise, it raisesan error.
[Procedure]im-rgba->rgb image [#:bg ’(0.0 0.0 0.0)][Procedure]im-rgba->gray image [#:bg ’(0.0 0.0 0.0)][Procedure]im-rgb->gray image
Returns a new RGB or GRAY image.
In the RGBA case, image channels are first normalized. The new RGB channels areobtained by applying the following pseudo code algorithm:
R = (((1 - Source.A) * BG.R) + (Source.A * Source.R)) * 255.0
G = (((1 - Source.A) * BG.G) + (Source.A * Source.G)) * 255.0
B = (((1 - Source.A) * BG.B) + (Source.A * Source.B)) * 255.0
[Procedure]im-local-minima image [#:threshold +float-max+][Procedure]im-local-maxima image [#:threshold (- +float-max+)][Procedure]im-local-minima-channel channel width height [#:threshold
+float-max+]
Chapter 3: API Reference 22
[Procedure]im-local-maxima-channel channel width height [#:threshold (-+float-max+)]All local mnima and maxima related procedures also accept the following addi-tional optional keyword arguments: [#:con 8] [#:marker 1.0] [#:borders? #f][#:plateaus? #f] [#:epsilon 1.0e-4]
Returns a new image or channel.
Finds the local minima or maxima in image or channel. Local minima or maxima aredefined as ‘points’ that are not on the borders (unless #:borders? is #t), and whosevalues are lower or higher, respectively, then the values of all direct neighbors. In theresult image or channel, these points are marked using the #:marker value (all otherpixels values will be set to 0).
By default, the algorithm uses 8-connectivity to define a neighborhood, which can bechanged passing the optional keyword argument #:con, which can be either 4 or 8.
The #:threshold optional keyword argument can be used to discard minima andmaxima whose (original pixel) value is not below or above the threshold, respectively.Both default values depend on +float-max+, which is defined (and so is +float-min+)using the corresponding limit value as given by the C float.h header.
The #:plateaus? optional keyword argument can be used to allow regions of‘constant’ (original pixel) value whose neighbors are all higher (minima) or lower(maxima) than the value of the region. Tow pixel values are considered part of thesame region (representing the same ‘constant’ value) if the absolute value of theirdifference is <= to #:epsilon.
Notes:
• If you want to know how many minima or maxima were found, use [im-reduce],page 24, upon the result;
• If you are interested by the original minima or maxima pixel values, Use [im-times], page 23, between the original image and the result.
[Procedure]im-threshold image threshold [#:bg ’black]Returns a new BLACK and WHITE image.
The image argument can either be a GRAY or an an RGB image, in which case eachpixel is converted to GRAY as it is processed. Valid #:bg values are black (thedefault) and white.
Pixels for which the original value is >= threshold are set to 255.0 if #:bg is ’black,and set to 0.0 if #:bg is ’white. The other pixels are set to 0.0 or 255.0 respectively.
[Method]im-add image val[Method]im-add i1 i2 i3 . . .[Method]im-add-channel channel width height val[Method]im-add-channel width height c1 c2 c3 . . .
Returns a new image or channel.
Performs the scalar addition of image with val or the matrix addition of i1 i2 i3 . . .or c1 c2 c3 . . . respectively.
Chapter 3: API Reference 23
[Method]im-subtract image val[Method]im-subtract i1 i2 i3 . . .[Method]im-subtract-channel channel width height val[Method]im-subtract-channel width height c1 c2 c3 . . .
Returns a new image or channel.
Performs the scalar subtraction of image with val or the matrix subtraction of i1 i2i3 . . . or c1 c2 c3 . . . respectively.
[Method]im-times image val[Method]im-times i1 i2 i3 . . .[Method]im-times-channel channel width height val[Method]im-times-channel c1 w1 h1 c2 w2 h2 c3 w3 h3 . . .
Returns a new image or channel.
Performs the scalar multiplication of image with val or the element by element mul-tiplication of i1 i2 i3 . . . or c1 c2 c3 . . . respectively.
[Procedure]im-mtimes i1 i2 i3 . . .[Procedure]im-mtimes-channel width height c1 c2 c3 . . .
Returns a new image or channel.
Performs matrix multiplication of i1 i2 i3 . . . or c1 w1 h1 c2 w2 h2 c3 w3 h3 . . .recursively. The number of lines of the next image must equal the number of columnsof the previous intermediate result.
[Method]im-divide image val[Method]im-divide i1 i2 i3 . . .[Method]im-divide-channel channel width height val[Method]im-divide-channel c1 w1 h1 c2 w2 h2 c3 w3 h3 . . .
Returns a new image or channel.
Performs the scalar division of image with val or the element by element division ofi1 i2 i3 . . . or c1 c2 c3 . . . respectively.
It is the user responsibility to insure that none of the c2 c3 . . . values is zero.
[Procedure]im-mdivide i1 i2 i3 . . .[Procedure]im-mdivide-channel width height c1 c2 c3 . . .
Returns a new image or channel.
Performs the matrix multiplication of i1 or c1 by the inverse of i2 i3 . . . or c2 c3 . . .recursively. The number of lines of the next image must equal the number of columnsof the previous intermediate result9.
It is the user responsibility to insure that none of the c2 c3 . . . values is zero.
[Procedure]im-range image[Procedure]im-range-channel channel width
Returns a list of six values (min row col max row col) if image is GRAY, and a listof list of these values if image is RGB or for any n-chan > 1 images.
9 Technically speaking, there is no such thing as matrix division. Dividing a matrix by another matrix is anundefined function. The closest equivalent is to multiply the matrix by the inverse of the other matrix.
Chapter 3: API Reference 24
[Procedure]im-min image[Procedure]im-max image[Procedure]im-min-channel channel width[Procedure]im-max-channel channel width
Returns three multiple values if image is GRAY: min row col or max row col respec-tively. If image is RGB or for any n-chan > 1 images, it returns a list of list of thesevalues.
[Procedure]im-map proc i1 i2 i3 . . .[Procedure]im-map-channel proc width height c1 c2 c3 . . .
Returns a new image or channel.
Apply proc to each pixel value of each channel of i1 (if only two arguments are given),or to the corresponding pixel values of each channels of i1 i2 i3 . . . (if more than twoarguments are given).
[Procedure]im-reduce image proc default[Procedure]im-reduce-channel channel proc default
Returns one value if image is GRAY. If image is RGB or for any n-chan > 1, it returnsa list of values.
If image is empty, im-reduce returns default (this is the only use for default). Ifimage has only one pixel, then the pixel value is what is returned. Otherwise, procis called on the pixel values of image.
Each proc call is (proc elem prev), where elem is a pixel value from the channel(the second and subsequent pixel values of the channel), and prev is the returnedvalue from the previous call to proc. The first pixel value - for each channel - is theprev for the first call to proc.
For example:
,use (cv)
(im-load "edx.png")
a$2 = (128 128 1 (#f32(4.0 26.0 102.0 97.0 58.0 10.0 9.0 21.0 # ...)))
(im-reduce $2 + 0)
a$3 = 556197.0
[Procedure]im-and i1 i2 i3 . . .[Procedure]im-and-channel width height c1 c2 c3 . . .[Procedure]im-or i1 i2 i3 . . .[Procedure]im-or-channel width height c1 c2 c3 . . .[Procedure]im-xor i1 i2 i3 . . .[Procedure]im-xor-channel width height c1 c2 c3 . . .
Returns image if one argument only, otherwise, it returns a new image or channel, asthe result of computing the logical AND, OR or XOR of all images or channels.
In the case of AND, for all positive results, the pixel values (of each channel) of thenew image are set to the one obtained from i1 or c1 respectively, and 0.0 otherwise.
In the case of OR, the pixel values (of each channel) of the new image are set to theone obtained from the first non zero ii or ci respectively, otherwise it is set to 0.0.
Chapter 3: API Reference 25
In the case of XOR, the pixel values (of each channel) of the new image are set to thevalue obtained from successively computing (logior (logand a (- 255 b)) (logand
(- 255 a) b)) where a would be the previous result and b the current image orchannel pixel value, until all images passed in arguments have been processed10.
All images must have the same width, height and n-channel.
There are, of course, scientific use and examples of images logical XOR, and that iswhy Guile-CV (http://www.gnu.org/software/guile-cv) is being developed for,but let’s have a bit of fun here, and see if our levitating GNU likes apples!
[Procedure]im-complement imageReturns a new image.
This procedure computes the mathematical complement of image, which for Guile-CVmeans that for each pixel of each channel, the new value is (- 255.0 pixel-value).
[Procedure]im-scrap image size [#:pred <] [#:con 8] [#:bg ’black][#:exclude-on-edges #f]
Returns a new image.
Scraping an image is the operation of removing objects depending on their size (inpixels). When exclude-on-edges is #t, all objects that are on any edges are alsoremoved.
The procedure first calls [im-label], page 36, using con and bg, then calls [im-features],page 26. The area feature of each object is then compared to size using pred andthe object is removed if the result is #t.
Note that image must be a binary image.
[Procedure]im-particles image features [#:clean #t]Returns two values, a list of images (the particles) and a list of their bounding boxesin the original image.
Each returned image is a ‘particle’, which is a subpart of image determined byits bounding box, given by the left top right bottom values of the corresponding‘entry’ in features (see [im-features], page 26, for a complete description of a featurevalue list.
When #:clean is #t, which is the default, [im-particle-clean], page 25, is called uponeach particle (see below for a description of the expected result).
[Procedure]im-particle-clean particleReturns a new image.
Cleaning a particle (which is an image) means detecting and removing any object(s)that is(are) not connected to the ‘particle’ itself.
This procedure is based on the property that in a ‘particle’, which is an (sub)imageresulting from a [im-crop], page 21, based on the bounding box coordinates as returnedby im-features, there precisely is one object that, if you call im-features uponparticle, would have its bounding box coordinates being the entire particle. In otherwords, if you call im-particle-clean upon an image that is not a ‘particle’, theresult will just be a black image.
10 Note that there is no mathematically valid XOR operation on floating points, hence as they are ‘accessed’,pixel values are converted to integer, using float->int, defined in the (cv support libguile-cv) module).
Chapter 3: API Reference 26
3.2.5 Features
The Guile-CV procedures and methods related to image features.
Procedures
[Procedure]im-features image l-image [#:n-label #f]Returns a list of features, one list for each labeled object - including the backgroud -in ascending order.
Notes: (a) image can either be an RGB or a GRAY image; (b) l-image is the‘corresponding’ labeled image; (c) when used, the #:n-label optional keyword argu-ment must be total number of label values used in l-image, as returned by [im-label],page 36, and [im-label-all], page 36.
The GRAY feature list values are:
area The labeled object area in pixel
left top right bottom
The coordinates of the ‘bounding box’ labeled object11
mean-x mean-y
Also sometimes called the ‘centroid’, these are the average ofthe x and y coordinates of all of the pixels in the labeled object.These two coordinate values are floating points, representing the‘mathematical position’ of the mean x and y values of the la-beled object
min max mean std-dev
The minimum, maximum, mean and standard gray deviaton la-beled object values
major-ev-x major-ev-y minor-ev-x minor-ev-y
Respectively the major and minor eigen vectors(https://en.wikipedia.org/wiki/Eigenvalues_and_eigenvectors)x and y normalized coordinates12: (= (sqrt (+ (expt x 2) (expt
y 2))) 1)
major-axis minor-axis
Respectively the major and minor eigen values(https://en.wikipedia.org/wiki/Eigenvalues_and_eigenvectors),optimized so that they actually correspond to major and minorradius of the ellipse covering the same area as the particle itself
angle The angle of the major eigen vector axis, in degrees in thetrigonometirc circle reference system
11 Note that when passed to [im-crop], page 21, right and bottom must be increased by 1: (im-crop image
left top (+ right 1) (+ bottom 1)).12 Note that Vigra calculates and returns these values in the image coordinate system, where the y-axis is
‘flipped’ compared to the trigonometric circle ‘traditional’ representation. Guile-CV however transformsand returns these values using the trigonometric circle reference system.
Chapter 3: API Reference 27
center-mass-x center-mass-y
The center of mass x and y coordinates
perimeter
The labeled object perimeter in pixels
skewness kurtosis
Respectively the skewness (https://en.wikipedia.org/wiki/Skewness)and the kurtosis (https://en.wikipedia.org/wiki/Kurtosis)of the labeled object
circularity aspect-ratio roundness
Respectively the circularity (/ (* 4 %pi area) (expt perimeter
2)), the aspect ratio (/ major-axis minor-axis) and the round-ness (/ minor-axis major-axis) of the labeled object
The RGB feature list values are:
area The labeled object area in pixel
left top right bottom
The coordinates of the labeled object (the corresponding GRAYfeature footnote applies here too of course)
mean-x mean-y
Also sometimes called the ‘centroid’, these are the average of thex and y coordinates of all of the (red green blue) pixels in thelabeled object. These two coordinate values are floating points,representing the ‘mathematical position’ of the mean x and yvalues of tha labeled object
min-r min-g min-b max-r max-g max-b mean-r mean-g mean-b std-dev-r
std-dev-g std-dev-b
The minimum, maximum, mean and standard deviaton labeled ob-ject values of the red, green and blue channels
major-axis minor-axis
Respectively the major and minor eigen values(https://en.wikipedia.org/wiki/Eigenvalues_and_eigenvectors),optimized so that they actually correspond to major and minorradius of the ellipse covering the same area as the particle itself
angle The angle of the major eigen vector axis, in degrees in thetrigonometirc circle reference system
center-mass-x center-mass-y
The center of mass x and y coordinates
perimeter
The labeled object perimeter in pixels
skewness-r skewness-g skewness-b kurtosis-r kurtosis-g kurtosis-b
Respectively the skewness (https://en.wikipedia.org/wiki/Skewness)and the kurtosis (https://en.wikipedia.org/wiki/Kurtosis)labeled object values of the red, green and blue channels
Chapter 3: API Reference 28
circularity aspect-ratio roundness
Respectively the circularity (/ (* 4 %pi area) (expt perimeter
2)), the aspect ratio (/ major-axis minor-axis) and the round-ness (/ minor-axis major-axis) of the labeled object
Though we did not make it public, Guile-CV has an internal feature display procedurethat you might be interested to (re)use, so here is an example of a GRAY feature listdisplay:
scheme@(guile-user)> ,use (cv)
scheme@(guile-user)> (im-load "pp-17-bf.png")
$2 = (85 95 3 (#f32(0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ...) ...))
scheme@(guile-user)> (im-rgb->gray $2)
$3 = (85 95 1 (#f32(0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 # ...)))
$4 = (0.0 251.0 128.3132714138286 8075)
scheme@(guile-user)> (im-threshold $3 136)
$5 = (85 95 1 (#f32(0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 # ...)))
scheme@(guile-user)> (im-label $5)
$6 = (85 95 1 (#f32(0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 # ...)))
$7 = 2
scheme@(guile-user)> (im-features $2 $6)
$8 = ((3782 0 0 84 94 43.19196319580078 45.657588958740234 0.0 # ...) ...)
scheme@(guile-user)> ((@@ (cv features) f-display) (match $8 ((bg a) a)))
area : 4293 (pixels)
left top right bottom : 0 0 84 94
mean-x mean-y : 40.94992 48.18262
min (red, green, blue) : 137.00000 136.00000 135.00000
max (red, green, blue) : 255.00000 250.00000 250.00000
mean (red, green, blue) : 236.13417 232.84999 232.84207
std. dev. (red, green, blue) : 20.23275 19.41402 19.84854
major ev x, y : 0.22202 0.97504
minor ev x, y : 0.97504 -0.22202
major, minor axis : 39.86419 34.27900 (radius)
angle : 77.17241 (degrees)
center of mass x, y : 40.73749 48.28692
perimeter : 367.74725
skewness (red, green, blue) : -2.90164 -2.99066 -2.91777
kurtosis (red, green, blue) : 8.53371 9.05482 8.61162
circularity : 0.39891
aspect ratio : 1.16293
roundness : 0.85989
As we mentioned above, f-diplay is defined in the (cv features) module, but itis not exported: in Guile, calling none exported procedure (which should not be‘abused’) is done using the syntax @@ module-name binding-name, which in this ex-ample translates to (@@ (cv features) f-display).
Chapter 3: API Reference 29
3.2.6 Texture
The Guile-CV procedures and methods related to image texture measures.
First described here13, image texture measures are still very ‘actual’, with new research andpracticle applications in many areas, as described in this (highly recommended) document14.
Image texture measures are ‘descriptive statistics’, derived from the ‘Gray Level
Co-occurrence Matrices (GLCM)’ and its associated set of ‘Gray Level Co-occurrence
Probability (GLCP)’ matrices.
Guile-CV GLCM and GLCP data structures are identical to the one used for Guile-CV images(See Section 3.2.1 [Image Structure and Accessors], page 10). Although they are not images‘per se’, they are composed of four square matrices (four channels), of size n-gl (thenumber of gray levels to consider), and upon which we (and users) need to run linearalgebra procedures, already defined and available in Guile-CV.
Guile-CV offers the 11th first texture measures, out of the 14th originally proposed byHaralick et al., which are the most commonly used and adopted ones.
This reference manual assumes you are familiar with the concepts, terminology and mathe-matic formulas involved in the calculations of GLCMs, GLCPs and image texture measures.If that is not the case, consider carefully reading one or both of the documents cited above(or any other tutorial or reference material of your choice of course).
Procedures
[Procedure]im-texture image n-gl [#:dist 1] [#:p-max 255] [#:use-log2 #f][#:no-px-y0 #f]
Returns a list.
The procedure calls [im-glcp], page 30, passing image, n-gl (the number of gray levelsto consider), dist (the distance between the ‘reference’ and the ‘neighbour’ pixels)and p-max (the image (pixel) maximum value), then computes and returns a list ofthe 11th first texture measures proposed by Haralick et al., which are:
(h1) uniformity (angular second moment)
(h2) contrast
(h3) correlation
(h4) variance (sum of squares)
(h5) homogeneity (inverse difference moment)
(h6) sum average
(h7) sum variance
(h8) sum entropy
(h9) entropy
(h10) difference variance
(h11) difference entropy
The #:use-log2 optional keyword argument, which defaults to #f, is passed to theinternal procedures that calculate the parameters h8, h9 and h11. The original formu-
13 R. M. Haralick, K. Shanmugam, and I. Dinstein, Textural Features of Image Classification, IEEE Transac-tions on Systems, Man and Cybernetics, vol. SMC-3, no. 6, Nov. 1973.
14 M. Hall-Beyer, GLCM Texture: A Tutorial v. 3.0 March 2017
Chapter 3: API Reference 30
las proposed by Haralck and al. use log, but I have seen a couple of implementationsusing log215.
The #:no-px-y0 optional keyword argument, which defaults to #f, is passed to theinternal procedure that calculate the parameter h10. For some obscure reason, andonly with respect to this parameter, I have seen some implementations eliminatingthe first element of the so-called Px-y, an internediate f32vector result, which holds,as its first element, the sum of the elements of the main diagnal of the GLCP16.
[Procedure]im-glcp image n-gl [#:dist 1] [#:p-max 255]Returns the GLCP for image.
The procedure calls [im-glcm], page 30, passing image, n-gl (the number of gray levelsto consider), dist (the distance between the ‘reference’ and the ‘neighbour’ pixels)and p-max (the image (pixel) maximum value), adds GLCM’ (the transposed versionof GLCM, so the result is symmetrical around the diagonal), then computes and returnsthe GLCP.
The returned GLCP is an ‘image’ composed four channels (four square matrices ofsize n-gl), corresponding to the (symmetrical) Gray Level Co-occurrences expressedas propabibilities, each calculated at a specific ‘angle’, respectively 0o, 45o, 90o, and135o.
[Procedure]im-glcm image n-gl [#:dist 1] [#:p-max 255]Returns the GLCM for image.
The procedure scales the original image (it brings its values in the range [0 (- n-gl
1)]), then computes and returns the GLCM.
The returned GLCM is an ‘image’ composed four channels (four square matrices of sizen-gl), corresponding to the Gray Level Co-occurrences, each calculated at a specific‘angle’, respectively 0o, 45o, 90o, and 135o.
3.2.7 Filters
The Guile-CV procedures and methods to filter images.
Procedures
[Procedure]im-gaussian-blur image sigma[Procedure]im-gaussian-blur-channel channel width height sigma
Returns a new image or channel.
The new image or new channel is the result of the computation of the Gaussianblurring, also known as the Gaussian smoothing, by means of a convolution of imageor channel with a 2D Gaussian function, where sigma is the standard deviation of theGaussian distribution.
15 Since it is used as a factor in all three formulas, the final result obtained using log2 is equivalent to theresult obtained using log multiplied by 1.4426950408889634
16 Guile-CV computes the difference average using all elements of the Px-y, by default, but offers thisoption as a courtesy, for users who would want to use Guile-CV as an immediate substitute to computeimage texture measures for a (large) image set for which they would already have trained a classifier. It isnot recommended to use it otherwise.
Chapter 3: API Reference 31
[Procedure]im-gaussian-gradient image sigma[Procedure]im-gaussian-gradient-channel channel width height sigma
Returns a new image or channel.
The new image or new channel is the result of the computation of the strength of thefirst order partial derivatives by means of a convolution of image or channel with thefirst order derivative of a 2D Gaussian function, where sigma is the standard deviationof the Gaussian distribution.
[Procedure]im-gaussian-sharp image factor scale[Procedure]im-gaussian-sharp-channel channel width height factor scale
Returns a new image or channel.
The new image or new channel is the result of the computation of the Gaussiansharpening: the procedure does (a) perform a Gaussian smoothing at the given scaleto create a temporary image smooth and (b) blends image and smooth according tothe formula (- (* (+ factor 1) image) (* smooth factor)).
[Procedure]im-sharpen image factor[Procedure]im-sharpen-channel channel width height factor
Returns a new image or channel.
This procedure performs a ‘simple sharpening’ operation on image. It actually calls[im-convolve], page 32, with the following kernel:
-1/16 -1/8 -1/16 0 0 0
( * factor -1/8 3/4 -1/8 ) + 0 1 0
-1/16 -1/8 -1/16 0 0 0
and uses mirror as the ‘out of bound strategy’.
[Procedure]im-median-filter image w-width w-height [#:obs ’repeat][Procedure]im-median-filter-channel channel width height w-width
w-height [#:obs ’repeat]Returns a new image or channel.
In the new image or channel, each pixel value is the ‘median’ value ofneighboring entries. The pattern of neighbors is called a ‘window’, thesize of which is given by w-width and w-height (see Median Filter(https://en.wikipedia.org/wiki/Median_filter) for more information).Both w-width and w-height must be odd numbers, inferior to width and heightrespectively.
The optional keyword argument #:obs determines the ‘out-of-bound strategy’.Valid #:obs symbols are:
avoid do not operate on pixels upon which (centering) the kernel does notfit in the image
repeat repeat the nearest pixels
mirror mirror the nearest pixels
wrap wrap image around (periodic boundary conditions)
zero out-of-bound pixel values to be 0.0
Chapter 3: API Reference 32
[Procedure]im-convolve image kernel [#:obs ’repeat][Procedure]im-convolve-channel channel width height kernel k-width k-height
[#:obs ’repeat]Returns a new image or channel.
The new image or new channel is the result of the convolution(https://en.wikipedia.org/wiki/Kernel_(image_processing)#Convolution)of image using kernel. The kernel (https://en.wikipedia.org/wiki/Kernel_(image_processing))k-width and k-height values can be different, but they must be odd numbers, inferiorto width and height respectively.
The optional keyword argument #:obs determines the ‘out-of-bound strategy’.Valid #:obs symbols are:
avoid do not operate on pixels upon which (centering) the kernel does notfit in the image
clip clip the kernel when operating on pixels upon which (centering) thekernel does not fit in the image (this is only useful if the kernel is>= 0 everywhere)
repeat repeat the nearest pixels
mirror mirror the nearest pixels
wrap wrap image around (periodic boundary conditions)
zero out-of-bound pixel values to be 0.0
Kernel data structure, accessors, procedures and predefined kernels are all describedin this node of the Guile-CV manual: Section 3.2.2 [Kernel Structure and Accessors],page 12.
[Procedure]im-nl-means image arg...[Procedure]im-nl-means-channel channel width height arg...
Returns a new image or channel.
The new image or new channel is the result of a non-local means(https://en.wikipedia.org/wiki/Non-local_means) denoising as de-scribed here17. The following table lists the optional keyword arguments and theirdefault values:
Policy arguments:
#:policy-type 1
accepts 0 (ratio policy) or 1 (norm policy)
#:sigma 15.0
default to 5.0 if the policy-type is 0
#:mean-ratio 5.0
default to 0.95 if the policy-type is 0
17 P. Coupe, P. Yger, S. Prima, P. Hellier, C. Kervrann, C. Barillot. An Optimized Blockwise Non Local MeansDenoising Filter for 3D Magnetic Resonance Images . IEEE Transactions on Medical Imaging, 27(4):425-441,Avril 2008.
Chapter 3: API Reference 33
#:variance-ratio 0.5
#:epsilon 1.0e-5
Filter arguments:
#:spatial-sigma 2.0
#:search-radius 3
#:patch-radius 1
the patch-radius can be either 1 or 2
#:mean-sigma 1.0
#:step-size 2
#:n-iteration 1
The im-nl-means-channel procedure accepts one additional optional keyword argu-ment:
#:n-thread (- (current-processor-count) 1)
FIXME need to describe the parameters
3.2.8 Morphology
The Guile-CV procedures and methods related to morphology.
Procedures
[Procedure]im-disc-erode image radius[Procedure]im-disc-erode-channel channel width height radius
Returns a new image or channel.
Performs the morpholgical erosion of image using a disc of a given radius. Here is anexample:
(im-make 5 5 1 1.0)
a$2 = (5 5 1 (#f32(1.0 1.0 1.0 1.0 1.0 ...)))
(im-set! $2 1 2 0.0)
(im-disc-erode $2 1)
a$3 = (5 5 1 (#f32(1.0 0.0 0.0 0.0 1.0 ...)))
(im-display $2 #:proc inexact->exact)
aChannel 1
1 1 1 1 1
1 1 0 1 1
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
(im-display $3 #:proc inexact->exact)
aChannel 1
1 0 0 0 1
1 0 0 0 1
Chapter 3: API Reference 34
1 0 0 0 1
1 1 1 1 1
1 1 1 1 1
[Procedure]im-disc-dilate image radius[Procedure]im-disc-dilate-channel channel width height radius
Returns a new image or channel.
Performs the morpholgical dilation of image using a disc of a given radius. Here isan example:
...
a$13 = (11 11 1 (#f32(0.0 0.0 0.0 0.0 0.0 ...)))
(im-disc-dilate $13 1)
a$14 = (11 11 1 (#f32(1.0 1.0 1.0 1.0 1.0 ...)))
(im-display $13 #:proc inexact->exact)
aChannel 1
0 0 0 0 0 0 0 0 0 0 0
0 1 1 1 1 0 0 1 1 1 0
0 1 1 1 1 0 0 1 1 1 0
0 1 1 1 1 1 1 1 1 1 0
0 1 1 1 1 1 1 1 1 1 0
0 1 1 0 0 0 1 1 1 1 0
0 1 1 0 0 0 1 1 1 1 0
0 1 1 0 0 0 1 1 1 1 0
0 1 1 1 1 1 1 1 0 0 0
0 1 1 1 1 1 1 1 0 0 0
0 0 0 0 0 0 0 0 0 0 0
(im-display $14 #:proc inexact->exact)
aChannel 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 0 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 0 0
1 1 1 1 1 1 1 1 1 0 0
[Procedure]im-open image radius[Procedure]im-open-channel channel width height radius
Returns a new image or channel.
Chapter 3: API Reference 35
Performs the dilation of the erosion of image using radius. Opening removes smallobjects.
[Procedure]im-close image radius[Procedure]im-close-channel channel width height radius
Returns a new image or channel.
Performs the erosion of the dilation of image using radius. Closing removes smallholes.
[Procedure]im-fill-holes image[Procedure]im-fill-holes-channel channel width height
Returns a new image or channel.
The argument must be a BINARY image. As its name indicate, this procedure fillthe holes of all and every objects in the image.
[Procedure]im-delineate image [#:threshold 10] [#:radius 2][Procedure]im-delineate-channel channel width height [#:threshold 10]
[#:radius 2]Returns a new image or channel.
Both threshold and radius must be exact integers.
Also known as ‘Edge Enhancement’, this procedure performs the delineation of image,obtained by applying the following pseudo code algorithm:
;; with
;; Min = (im-disc-erode image radius)
;; Max = (im-disc-dilate image radius)
D = Max - Min
If D < threshold
;; not an edge
output pixel = input pixel
;; it is an edge
If (pixel -- Min) < (Max -- pixel)
output pixel = Min
output pixel = Max
Here above, left being the original image - a small part of an optical microscopecapture of a sinter sample - you can see the difference between im-delineate calledwith the default threshold and radius values, then called using #:threshold 25 and#:radius 5.
[Procedure]im-distance-map image [#:bg ’black] [#:mode ’euclidean][Procedure]im-distance-map-channel channel width height [#:bg ’black]
[#:mode ’euclidean]Returns a new image or channel.
Also know as ‘Distance Tranform’, this procedure performs the distance map ofimage, which consist of, for each background pixel, calculating its distance to thenearest object or contour. In the return new image or channel, all background pix-els will be assigned the their distance value, all other pixels will be assigned to 0.Distances larger than 255 are labelled as 255.
Chapter 3: API Reference 36
The default backgroung pixel value is ’black, the optional #:bg keyword argumentalso accepts ’white.
The default distance map mode is ’euclidean (https://en.wikipedia.org/wiki/Euclidean_distance).Other valid optional #:mode keyword argument are ’chessboard(https://en.wikipedia.org/wiki/Chessboard_distance) and ’manhattan(https://en.wikipedia.org/wiki/Taxicab_geometry).
Here above, left being the original image - a few cells - you can see the results obtainedby calling im-distance-map using respectively the ’euclidean, ’manhattan and’chessboard modes.
[Procedure]im-reconstruct image seeds [#:con 8]Returns a new image.
This procedure implements a ‘binary morphological reconstruction’ algorithm,which extracts the connected components of image that are ‘marked’ by (any of) theconnected components contained in seeds.
Morphological reconstruction is part of a set of image operators often referred to as‘geodesic’ (geodesic distance, geodesic dilation . . . ). Morphological (or geodesic)operations upon digital images come from and use the Mathematical morphology(MM) (https://en.wikipedia.org/wiki/Mathematical_morphology) theory andtechnique developed for the analysis and processing of geometrical structures.
First described here18, this implementation is based on a revisionof the same article published in ‘the IEEE Transactions on Image
Processing, Vol. 2, No. 2, pp. 176-201, April 1993’, available here(http://www.vincent-net.com/luc/papers/93ieeeip_recons.pdf).
3.2.9 Segmentation
The Guile-CV procedures and methods related to segmentation.
Procedures
[Procedure]im-label image [#:con 8] [#:bg ’black][Procedure]im-label-channel channel width height [#:con 8] [#:bg ’black][Procedure]im-label-all image [#:con 8][Procedure]im-label-all-channel channel width height [#:con 8]
Returns two values: a new GRAY image or channel, and the total number of labels19.
The im-label and im-label-channel procedures label foreground objects in thebinary image. In the new image or channel, 0.0 indicates a background pixel, 1.0indicates that the pixel belongs to object number 1, 2.0 that the pixel belongs toobject number 2, etc.
18 in Serra, Jean and Vincent, Luc (1992), "An overview of morphological filtering", Circuits, Systems andSignal Processing (Springer) 11 (1): 47-108
19 The number of labels correspond to the highest label value + 1: earlier version of Guile-CV, prior to version1.8.0, did return the number of objects, which correspond to the highest label value. This was less thenoptimal, since not only 0.0 is a label, but other procedures, im-features for example, do consider and returnand element for the background as well.
Chapter 3: API Reference 37
The im-label-all and im-label-all-channel procedures label all objects in thebinary image, with no specific distinction for any background value. As a result, thesetwo procedures will label not only the continuous background, if any, but also anyhole(s). As an example, they are used by [im-fill-holes], page 35, defined in the module(cv morphology), which you may have a look at for a better understanding of howit works.
Two pixels belong to the same object if they are neighbors. By default the algorithmuses 8-connectivity to define a neighborhood, but this can be changed through thekeyword argument #:con, which can be either 4 or 8.
[Procedure]im-canny image [#:sigma 1.0] [#:threshold 0.0] [#:marker 255.0][Procedure]im-canny-channel channel width height [#:sigma 1.0] [#:threshold
0.0] [#:marker 255.0]Returns a new image or channel.
Detect and mark edges using a Canny Edge Detector (https://en.wikipedia.org/wiki/Canny_edge_detector)algorithm: (a) compute the image Gaussian gradient using sigma,(b) remove edges whose strength is below threshold, then for all re-maining edges, (d) remove the non-local maxima (edge thinning(https://en.wikipedia.org/wiki/Edge_detection#Edge_thinning)) and(e) set their intensity using marker.
[Procedure]im-crack-edge image [#:marker 255.0][Procedure]im-crack-edge-channel channel width height [#:marker 255.0]
Returns a new image or channel.
Crack edges are marked ‘between’ the (different) pixels of image. In order to accom-modate the cracks, the resulting image or channel size must be (* (- width 1) 2) and(* (- height 1) 2) respectively.
Crack pixels are first inserted, then all crack pixels whose non-crack neighbors havedifferent values are crack edges and marked using marker, while all other pixels (crackand non-crack) become region pixels. Here is a simple example, with two regions, aand b, and using * as the crack edge marker:
Original Inserted Cracks Final Result
a b b
a a b
a a a
a . b . b
. . . . .
a . a . b
. . . . .
a . a . a
a * b b b
a * * * b
a a a * b
a a a * *
a a a a a
Crack Edge Images have the following properties:
• Crack Edge Images have odd width and height.
• Crack pixels have at least one odd coordinate.
• Only crack pixels may be marked as crack edge pixels.
• Crack pixels with two odd coordinates must be marked as edge pixels wheneverany of their neighboring crack pixels was marked.
Chapter 3: API Reference 38
As a consequence of the last two properties, both edges and regions are 4-connected.Thus, 4-connectivity and 8-connectivity yield identical connected components inCrack Edge Images (the so called well-composedness). This ensures that Crack EdgeImages have nice topological properties20.
3.2.10 Utilities
Other Guile-CV procedures and methods utilities.
Procedures
[Procedure]im-compose position alignment [#:color ’(0 0 0)] img-1 img-2 . . .[Procedure]im-compose-channels position alignment channels widths heights
[#:value ’0.0]Returns a new image or a new channel.
The valid position and alignment symbols are:
left right
top center bottom
above below
left center right
When used, the optional #:color keyword argument must come after the mandatoryalignment argument and precede img-1, otherwise Guile will raise an exception. ForRGB images, it is the color used to padd images passed in argument that are inferior,in width or height (depending on the position), to the biggest of them. For GRAYimages, the #:color is reduced to its corresponding gray value:
(/ (reduce + 0 color) 3)
For the im-compose-channels procedure, the list of channels, widths and heightsmust be of equal length and equally ordered, so the nth element of widths and heightsare the width and height of the nth element of channels. The optional #:valuekeyword argument is used to padd channels that are inferior, in width or height(depending on the position), to the biggest of them.
[Procedure]im-display image [#:proc #f] [#:port (current-output-port)][Procedure]im-display-channel channel width height [#:proc #f] [#:port
(current-output-port)]Returns nothing.
Displays the content of image or channel on port.
The optional #:proc keyword argument must either be #f, the default, or a procedurethat accepts a single (32 bits float) argument. When #:proc is #f, im-display willuse an internally defined procedure which formats its argument ‘a la octave’: ninepositions, six decimals, all number aligned on the dot. any value >= 1000 is convertedto use the exponential float notation. Here is an ‘hand made’ example:
...
$2 = (4 3 3 (#f32(0.0 1.0 2.0 3.0 4.0 5.0) ... ...)
20 See L. J. Latecki: Well-Composed Sets, Academic Press, 2000
Chapter 3: API Reference 39
scheme@(guile-user)> (im-divide $2 99)
$3 = (4 3 3 (#f32(10.1010103225708 0.010101010091602802 ...) ...))
scheme@(guile-user)> (im-set! $3 0 0 0 10000)
$4 = (4 3 3 (#f32(10000.0 0.010101010091602802 # # # # ...) ...))
scheme@(guile-user)> (im-display $4)
a
Channel 1
1.0E+4 0.01010 0.02020 0.03030
0.04040 0.05051 0.06061 0.07071
0.08081 0.09091 0.10101 0.11111
Channel 2
0.12121 0.13131 0.14141 0.15152
0.16162 0.17172 0.18182 0.19192
0.20202 0.21212 0.22222 0.23232
Channel 3
0.24242 0.25253 0.26263 0.27273
0.28283 0.29293 0.30303 0.31313
0.32323 0.33333 0.34343 0.35354
Caution: unless you specify port, both this and [im-display-channel], page 38, pro-cedures are ment to be used on very small and testing images, otherwise even on asmall image, it might be ok in a terminal, but it will definitely will kill your emacs.
[Procedure]im-histogram image [#:subtitle “Untitled”]Returns two values: (1) an image; (2) a list (or a list of list) of significant values forimage: one list if image is GRAY, a list of list of values per channel if image is RGB.
The returned image is composed of a header (title, #:subtitle), either the GRAY orthe RGB channel histogram(s) for image and a footer, which is a table containg, foreach channel, the following values: mean, standard deviation, minimum, maximum,the mode21 followed by its value.
Here below, the call sequence and the histogram for the GRAY image sinter.png
given along with Guile-CV documentation and examples:
scheme@(guile-user)> (im-load "sinter.png")
$32 = (212 128 1 (#f32(25.0 39.0 50.0 52.0 51.0 45.0 # ...)))
scheme@(guile-user)> (im-histogram $32 #:subtitle "sinter.png")
$34 = (282 271 1 (#f32(255.0 255.0 255.0 255.0 255.0 # ...)))
$35 = (27136 163.346 75.081 0 243 215 727)
Note that histogram images returned by im-histogram have no borders, the abovehistogram has been padded - using (im-padd $34 1 1 1 1 #:color ’(96 96 96)) - for
21 The mode is the integer corresponding to the histogram entry that received the maximum of hits, and thevalue displayed in parens precisely is the number of hits.
Chapter 3: API Reference 40
better readability, otherwise the title above and the table below would look as if theywere not centered.
[Method]im-show filename[Method]im-show image [scale #f][Method]im-show image name [scale #f]
Returns the string "#<Image: . . .>", where ". . ." is either filename or a filenameconstructed by im-show, see below.
The optional scale argument can take the following values:
#f pixel values are ‘clipped’: values < 0 are saved as 0, values > 255
are saved as 255, and otherwise are saved unchanged
#t all pixel values are scaled22 to the [0 255] range
These three methods will also effectively dislay the image if you are using Geiser(http://www.nongnu.org/geiser), which analyzes Guile’s procedures and methodsreturned values (through the use of its pattern matcher), and when appropriate,triggers its image display mechanism.
Geiser has two variables that allow you to choose either to inline images in its Emacs(https://www.gnu.org/software/emacs) (Guile repl) buffer, or to display themusing externel viewer: geiser-image-viewer and geiser-repl-inline-images-p.You may choose to add these variables in your .emacs file, for example:
(setq geiser-image-viewer "eog")
(setq geiser-repl-inline-images-p nil)
Note that (setq geiser-repl-inline-images-p t) will only work if you are usinga graphics-aware Emacs, and otherwise, will fall on the external viewer approach, ifthe variable geiser-image-viewer has been defined. When using Geiser in a nongraphics-aware Emac, or when using the external viewer approach, images will appearas buttons: press return on them to invoke (or raise) the external viewer (windowcontaining that image).
Except for the first im-show method, Guile-CV has to save the image first, and does itin the location defined by the [%image-cache], page 41, variable. If you call im-showpassing name, the image is saved as %image-cache/name.png, otherwise under agenerated name, the result of (symbol->string (gensym "im-show-")).
Note that if you do not specify name, a new external viewer window is opened ateach im-show invocation, even for identical image calls: this because in Guile-CV,on purpose, images are just list, with no (unique) identifier, and there is no way forim-show to know ... Further to this point, when you pass name as an argument, youare not ‘identifying’ image, which may actually differ, but rather just ask to reusethe filename and hence the external viewer window associated with it.
Last note: many external viewers, such as Eog (the Gnome Eye Viewer), will tryto apply, per default, some sort of smoothing techniques, especially on zoom-in andzoom-out: where this is fine for viewing ‘lazer’ pictures, you probably want to checkand disable these options when working with Guile-CV.
22 Note that in this particular context, scale does not mean a change in dimension, but rather bringing pixelvalues from the range they occupy in memory to the [0 255] range
Chapter 3: API Reference 41
Variables
[Variable]%image-cacheSpecifies the location used by [im-show], page 40, to save images.
The default value is /tmp/<username>/guile-cv, but you may set! it. If you’d liketo reuse that location for future guile-cv sessions, you may save it in guile-cv’s ‘peruser’ config file <userdir>/.config/guile-cv as an assoc pair, here is an example:
cat ~/.config/guile-cv.conf
((image-cache . "~/tmp"))
Note that if used, the ‘~’ is expanded at load time, so in geiser, it becomes:
scheme@(guile-user)> ,use (cv)
scheme@(guile-user)> %image-cache
a$2 = "/home/david/tmp"
[Variable]%image-cache-formatSpecifies the format used by [im-show], page 40, to save images.
The default value is "png", but you may set! it. If you’d like to reuse that for-mat for future guile-cv sessions, you may save it in guile-cv’s ‘per user’ config file<userdir>/.config/guile-cv, as an assoc pair, here is an example:
cat ~/.config/guile-cv.conf
((image-cache-format . "jpg"))
3.3 Support
Guile-CV uses a series of support modules, each documented in the following subsections.You may either import them all, like this (use-modules (cv support)), or individually,such as (use-modules (cv support modules)), (use-modules (cv support goops)), ...
3.3.1 Modules
[re-export-public-interface], page 41
[Special Form]re-export-public-interface . argsRe-export the public interface of a module or modules. Invoked as (re-export-
modules (mod1) (mod2)...).
3.3.2 Goops
3.3.3 G-export
3.3.4 Pi
Procedures
[Procedure]radian->degree rad[Procedure]degree->radian deg
Returns respectively a degree or a radian value.
Appendix A: GNU Free Documentation License 42
Variables
[Variable]%pi[Variable]%2pi[Variable]%pi/2
Respectively bound to (acos -1), (* 2 %pi) and (/ %pi 2).
3.3.5 Utils
Appendix A GNU Free Documentation License
Copyright c© 2000, 2001, 2002, 2007, 2008 Free Software Foundation, Inc.http://fsf.org/
Everyone is permitted to copy and distribute verbatim copiesof this license document, but changing it is not allowed.
0. PREAMBLE
The purpose of this License is to make a manual, textbook, or other functional anduseful document free in the sense of freedom: to assure everyone the effective freedomto copy and redistribute it, with or without modifying it, either commercially or non-commercially. Secondarily, this License preserves for the author and publisher a wayto get credit for their work, while not being considered responsible for modificationsmade by others.
This License is a kind of “copyleft”, which means that derivative works of the documentmust themselves be free in the same sense. It complements the GNU General PublicLicense, which is a copyleft license designed for free software.
We have designed this License in order to use it for manuals for free software, becausefree software needs free documentation: a free program should come with manualsproviding the same freedoms that the software does. But this License is not limited tosoftware manuals; it can be used for any textual work, regardless of subject matter orwhether it is published as a printed book. We recommend this License principally forworks whose purpose is instruction or reference.
1. APPLICABILITY AND DEFINITIONS
This License applies to any manual or other work, in any medium, that contains anotice placed by the copyright holder saying it can be distributed under the termsof this License. Such a notice grants a world-wide, royalty-free license, unlimited induration, to use that work under the conditions stated herein. The “Document”,below, refers to any such manual or work. Any member of the public is a licensee, andis addressed as “you”. You accept the license if you copy, modify or distribute the workin a way requiring permission under copyright law.
A “Modified Version” of the Document means any work containing the Document ora portion of it, either copied verbatim, or with modifications and/or translated intoanother language.
A “Secondary Section” is a named appendix or a front-matter section of the Documentthat deals exclusively with the relationship of the publishers or authors of the Document
Appendix A: GNU Free Documentation License 43
to the Document’s overall subject (or to related matters) and contains nothing thatcould fall directly within that overall subject. (Thus, if the Document is in part atextbook of mathematics, a Secondary Section may not explain any mathematics.) Therelationship could be a matter of historical connection with the subject or with relatedmatters, or of legal, commercial, philosophical, ethical or political position regardingthem.
The “Invariant Sections” are certain Secondary Sections whose titles are designated, asbeing those of Invariant Sections, in the notice that says that the Document is releasedunder this License. If a section does not fit the above definition of Secondary then it isnot allowed to be designated as Invariant. The Document may contain zero InvariantSections. If the Document does not identify any Invariant Sections then there are none.
The “Cover Texts” are certain short passages of text that are listed, as Front-CoverTexts or Back-Cover Texts, in the notice that says that the Document is released underthis License. A Front-Cover Text may be at most 5 words, and a Back-Cover Text maybe at most 25 words.
A “Transparent” copy of the Document means a machine-readable copy, representedin a format whose specification is available to the general public, that is suitable forrevising the document straightforwardly with generic text editors or (for images com-posed of pixels) generic paint programs or (for drawings) some widely available drawingeditor, and that is suitable for input to text formatters or for automatic translation toa variety of formats suitable for input to text formatters. A copy made in an otherwiseTransparent file format whose markup, or absence of markup, has been arranged tothwart or discourage subsequent modification by readers is not Transparent. An imageformat is not Transparent if used for any substantial amount of text. A copy that isnot “Transparent” is called “Opaque”.
Examples of suitable formats for Transparent copies include plain ascii withoutmarkup, Texinfo input format, LaTEX input format, SGML or XML using a publiclyavailable DTD, and standard-conforming simple HTML, PostScript or PDF designedfor human modification. Examples of transparent image formats include PNG, XCF
and JPG. Opaque formats include proprietary formats that can be read and editedonly by proprietary word processors, SGML or XML for which the DTD and/orprocessing tools are not generally available, and the machine-generated HTML,PostScript or PDF produced by some word processors for output purposes only.
The “Title Page” means, for a printed book, the title page itself, plus such followingpages as are needed to hold, legibly, the material this License requires to appear in thetitle page. For works in formats which do not have any title page as such, “Title Page”means the text near the most prominent appearance of the work’s title, preceding thebeginning of the body of the text.
The “publisher” means any person or entity that distributes copies of the Documentto the public.
A section “Entitled XYZ” means a named subunit of the Document whose title eitheris precisely XYZ or contains XYZ in parentheses following text that translates XYZ inanother language. (Here XYZ stands for a specific section name mentioned below, suchas “Acknowledgements”, “Dedications”, “Endorsements”, or “History”.) To “Preserve
Appendix A: GNU Free Documentation License 44
the Title” of such a section when you modify the Document means that it remains asection “Entitled XYZ” according to this definition.
The Document may include Warranty Disclaimers next to the notice which states thatthis License applies to the Document. These Warranty Disclaimers are considered tobe included by reference in this License, but only as regards disclaiming warranties:any other implication that these Warranty Disclaimers may have is void and has noeffect on the meaning of this License.
2. VERBATIM COPYING
You may copy and distribute the Document in any medium, either commercially ornoncommercially, provided that this License, the copyright notices, and the licensenotice saying this License applies to the Document are reproduced in all copies, andthat you add no other conditions whatsoever to those of this License. You may not usetechnical measures to obstruct or control the reading or further copying of the copiesyou make or distribute. However, you may accept compensation in exchange for copies.If you distribute a large enough number of copies you must also follow the conditionsin section 3.
You may also lend copies, under the same conditions stated above, and you may publiclydisplay copies.
3. COPYING IN QUANTITY
If you publish printed copies (or copies in media that commonly have printed covers) ofthe Document, numbering more than 100, and the Document’s license notice requiresCover Texts, you must enclose the copies in covers that carry, clearly and legibly, allthese Cover Texts: Front-Cover Texts on the front cover, and Back-Cover Texts onthe back cover. Both covers must also clearly and legibly identify you as the publisherof these copies. The front cover must present the full title with all words of the titleequally prominent and visible. You may add other material on the covers in addition.Copying with changes limited to the covers, as long as they preserve the title of theDocument and satisfy these conditions, can be treated as verbatim copying in otherrespects.
If the required texts for either cover are too voluminous to fit legibly, you should putthe first ones listed (as many as fit reasonably) on the actual cover, and continue therest onto adjacent pages.
If you publish or distribute Opaque copies of the Document numbering more than 100,you must either include a machine-readable Transparent copy along with each Opaquecopy, or state in or with each Opaque copy a computer-network location from whichthe general network-using public has access to download using public-standard networkprotocols a complete Transparent copy of the Document, free of added material. Ifyou use the latter option, you must take reasonably prudent steps, when you begindistribution of Opaque copies in quantity, to ensure that this Transparent copy willremain thus accessible at the stated location until at least one year after the last timeyou distribute an Opaque copy (directly or through your agents or retailers) of thatedition to the public.
It is requested, but not required, that you contact the authors of the Document wellbefore redistributing any large number of copies, to give them a chance to provide youwith an updated version of the Document.
Appendix A: GNU Free Documentation License 45
4. MODIFICATIONS
You may copy and distribute a Modified Version of the Document under the conditionsof sections 2 and 3 above, provided that you release the Modified Version under preciselythis License, with the Modified Version filling the role of the Document, thus licensingdistribution and modification of the Modified Version to whoever possesses a copy ofit. In addition, you must do these things in the Modified Version:
A. Use in the Title Page (and on the covers, if any) a title distinct from that of theDocument, and from those of previous versions (which should, if there were any,be listed in the History section of the Document). You may use the same title asa previous version if the original publisher of that version gives permission.
B. List on the Title Page, as authors, one or more persons or entities responsible forauthorship of the modifications in the Modified Version, together with at least fiveof the principal authors of the Document (all of its principal authors, if it has fewerthan five), unless they release you from this requirement.
C. State on the Title page the name of the publisher of the Modified Version, as thepublisher.
D. Preserve all the copyright notices of the Document.
E. Add an appropriate copyright notice for your modifications adjacent to the othercopyright notices.
F. Include, immediately after the copyright notices, a license notice giving the publicpermission to use the Modified Version under the terms of this License, in the formshown in the Addendum below.
G. Preserve in that license notice the full lists of Invariant Sections and required CoverTexts given in the Document’s license notice.
H. Include an unaltered copy of this License.
I. Preserve the section Entitled “History”, Preserve its Title, and add to it an itemstating at least the title, year, new authors, and publisher of the Modified Versionas given on the Title Page. If there is no section Entitled “History” in the Docu-ment, create one stating the title, year, authors, and publisher of the Documentas given on its Title Page, then add an item describing the Modified Version asstated in the previous sentence.
J. Preserve the network location, if any, given in the Document for public access toa Transparent copy of the Document, and likewise the network locations given inthe Document for previous versions it was based on. These may be placed in the“History” section. You may omit a network location for a work that was publishedat least four years before the Document itself, or if the original publisher of theversion it refers to gives permission.
K. For any section Entitled “Acknowledgements” or “Dedications”, Preserve the Titleof the section, and preserve in the section all the substance and tone of each of thecontributor acknowledgements and/or dedications given therein.
L. Preserve all the Invariant Sections of the Document, unaltered in their text andin their titles. Section numbers or the equivalent are not considered part of thesection titles.
Appendix A: GNU Free Documentation License 46
M. Delete any section Entitled “Endorsements”. Such a section may not be includedin the Modified Version.
N. Do not retitle any existing section to be Entitled “Endorsements” or to conflict intitle with any Invariant Section.
O. Preserve any Warranty Disclaimers.
If the Modified Version includes new front-matter sections or appendices that qualifyas Secondary Sections and contain no material copied from the Document, you may atyour option designate some or all of these sections as invariant. To do this, add theirtitles to the list of Invariant Sections in the Modified Version’s license notice. Thesetitles must be distinct from any other section titles.
You may add a section Entitled “Endorsements”, provided it contains nothing butendorsements of your Modified Version by various parties—for example, statements ofpeer review or that the text has been approved by an organization as the authoritativedefinition of a standard.
You may add a passage of up to five words as a Front-Cover Text, and a passage of upto 25 words as a Back-Cover Text, to the end of the list of Cover Texts in the ModifiedVersion. Only one passage of Front-Cover Text and one of Back-Cover Text may beadded by (or through arrangements made by) any one entity. If the Document alreadyincludes a cover text for the same cover, previously added by you or by arrangementmade by the same entity you are acting on behalf of, you may not add another; butyou may replace the old one, on explicit permission from the previous publisher thatadded the old one.
The author(s) and publisher(s) of the Document do not by this License give permissionto use their names for publicity for or to assert or imply endorsement of any ModifiedVersion.
5. COMBINING DOCUMENTS
You may combine the Document with other documents released under this License,under the terms defined in section 4 above for modified versions, provided that youinclude in the combination all of the Invariant Sections of all of the original documents,unmodified, and list them all as Invariant Sections of your combined work in its licensenotice, and that you preserve all their Warranty Disclaimers.
The combined work need only contain one copy of this License, and multiple identicalInvariant Sections may be replaced with a single copy. If there are multiple InvariantSections with the same name but different contents, make the title of each such sectionunique by adding at the end of it, in parentheses, the name of the original author orpublisher of that section if known, or else a unique number. Make the same adjustmentto the section titles in the list of Invariant Sections in the license notice of the combinedwork.
In the combination, you must combine any sections Entitled “History” in the vari-ous original documents, forming one section Entitled “History”; likewise combine anysections Entitled “Acknowledgements”, and any sections Entitled “Dedications”. Youmust delete all sections Entitled “Endorsements.”
6. COLLECTIONS OF DOCUMENTS
You may make a collection consisting of the Document and other documents releasedunder this License, and replace the individual copies of this License in the various
Appendix A: GNU Free Documentation License 47
documents with a single copy that is included in the collection, provided that youfollow the rules of this License for verbatim copying of each of the documents in allother respects.
You may extract a single document from such a collection, and distribute it individu-ally under this License, provided you insert a copy of this License into the extracteddocument, and follow this License in all other respects regarding verbatim copying ofthat document.
7. AGGREGATION WITH INDEPENDENT WORKS
A compilation of the Document or its derivatives with other separate and independentdocuments or works, in or on a volume of a storage or distribution medium, is calledan “aggregate” if the copyright resulting from the compilation is not used to limit thelegal rights of the compilation’s users beyond what the individual works permit. Whenthe Document is included in an aggregate, this License does not apply to the otherworks in the aggregate which are not themselves derivative works of the Document.
If the Cover Text requirement of section 3 is applicable to these copies of the Document,then if the Document is less than one half of the entire aggregate, the Document’s CoverTexts may be placed on covers that bracket the Document within the aggregate, or theelectronic equivalent of covers if the Document is in electronic form. Otherwise theymust appear on printed covers that bracket the whole aggregate.
8. TRANSLATION
Translation is considered a kind of modification, so you may distribute translationsof the Document under the terms of section 4. Replacing Invariant Sections withtranslations requires special permission from their copyright holders, but you mayinclude translations of some or all Invariant Sections in addition to the original versionsof these Invariant Sections. You may include a translation of this License, and all thelicense notices in the Document, and any Warranty Disclaimers, provided that youalso include the original English version of this License and the original versions ofthose notices and disclaimers. In case of a disagreement between the translation andthe original version of this License or a notice or disclaimer, the original version willprevail.
If a section in the Document is Entitled “Acknowledgements”, “Dedications”, or “His-tory”, the requirement (section 4) to Preserve its Title (section 1) will typically requirechanging the actual title.
9. TERMINATION
You may not copy, modify, sublicense, or distribute the Document except as expresslyprovided under this License. Any attempt otherwise to copy, modify, sublicense, ordistribute it is void, and will automatically terminate your rights under this License.
However, if you cease all violation of this License, then your license from a particularcopyright holder is reinstated (a) provisionally, unless and until the copyright holderexplicitly and finally terminates your license, and (b) permanently, if the copyrightholder fails to notify you of the violation by some reasonable means prior to 60 daysafter the cessation.
Moreover, your license from a particular copyright holder is reinstated permanently ifthe copyright holder notifies you of the violation by some reasonable means, this is thefirst time you have received notice of violation of this License (for any work) from that
Appendix A: GNU Free Documentation License 48
copyright holder, and you cure the violation prior to 30 days after your receipt of thenotice.
Termination of your rights under this section does not terminate the licenses of partieswho have received copies or rights from you under this License. If your rights havebeen terminated and not permanently reinstated, receipt of a copy of some or all of thesame material does not give you any rights to use it.
10. FUTURE REVISIONS OF THIS LICENSE
The Free Software Foundation may publish new, revised versions of the GNU FreeDocumentation License from time to time. Such new versions will be similar in spiritto the present version, but may differ in detail to address new problems or concerns.See http://www.gnu.org/copyleft/.
Each version of the License is given a distinguishing version number. If the Documentspecifies that a particular numbered version of this License “or any later version”applies to it, you have the option of following the terms and conditions either of thatspecified version or of any later version that has been published (not as a draft) bythe Free Software Foundation. If the Document does not specify a version number ofthis License, you may choose any version ever published (not as a draft) by the FreeSoftware Foundation. If the Document specifies that a proxy can decide which futureversions of this License can be used, that proxy’s public statement of acceptance of aversion permanently authorizes you to choose that version for the Document.
11. RELICENSING
“Massive Multiauthor Collaboration Site” (or “MMC Site”) means any World WideWeb server that publishes copyrightable works and also provides prominent facilitiesfor anybody to edit those works. A public wiki that anybody can edit is an example ofsuch a server. A “Massive Multiauthor Collaboration” (or “MMC”) contained in thesite means any set of copyrightable works thus published on the MMC site.
“CC-BY-SA” means the Creative Commons Attribution-Share Alike 3.0 license pub-lished by Creative Commons Corporation, a not-for-profit corporation with a principalplace of business in San Francisco, California, as well as future copyleft versions of thatlicense published by that same organization.
“Incorporate” means to publish or republish a Document, in whole or in part, as partof another Document.
An MMC is “eligible for relicensing” if it is licensed under this License, and if all worksthat were first published under this License somewhere other than this MMC, andsubsequently incorporated in whole or in part into the MMC, (1) had no cover textsor invariant sections, and (2) were thus incorporated prior to November 1, 2008.
The operator of an MMC Site may republish an MMC contained in the site underCC-BY-SA on the same site at any time before August 1, 2009, provided the MMC iseligible for relicensing.
Appendix A: GNU Free Documentation License 49
ADDENDUM: How to use this License for your documents
To use this License in a document you have written, include a copy of the License in thedocument and put the following copyright and license notices just after the title page:
Copyright (C) year your name.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License, Version 1.3
or any later version published by the Free Software Foundation;
with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
Texts. A copy of the license is included in the section entitled ‘‘GNU
Free Documentation License’’.
If you have Invariant Sections, Front-Cover Texts and Back-Cover Texts, replace the“with. . .Texts.” line with this:
with the Invariant Sections being list their titles, with
the Front-Cover Texts being list, and with the Back-Cover Texts
being list.
If you have Invariant Sections without Cover Texts, or some other combination of the three,merge those two alternatives to suit the situation.
If your document contains nontrivial examples of program code, we recommend releasingthese examples in parallel under your choice of free software license, such as the GNUGeneral Public License, to permit their use in free software.
50
Concept Index
This index contains concepts, keywords and non-Schemey names for several features, tomake it easier to locate the desired sections.
CCopying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Crack Edge Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
DDistance Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35Distance Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
EEdge Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37Edge Enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35Euclidean Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
FFeatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
GGaussian Blur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30Gaussian Gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Gaussian Sharp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Gaussian Smooth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30Geodesic Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36GLCM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30GLCP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30GPL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Gray Level Co-occurence Matrix . . . . . . . . . . . . . . . 30
Gray Level Co-occurence Probability . . . . . . . . . . . 30
IImage Convolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32Image Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25Image Textures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
LLicense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
MMathematical Morphology . . . . . . . . . . . . . . . . . . . . . 36Matrix Division . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23Matrix Multiplication . . . . . . . . . . . . . . . . . . . . . . . . . . 23Median Filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Morphological Reconstruction . . . . . . . . . . . . . . . . . . 36
NNon-Local Means Denoising . . . . . . . . . . . . . . . . . . . . 32
UUser Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
51
Procedure Index
This is an alphabetical list of all the procedures, methods and macros in Guile-CV.
Ddegree->radian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Iim-=-channel? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-=? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-add . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22im-add-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22im-and . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-and-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-binary-channel? . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-binary? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-canny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37im-canny-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37im-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-channel-offset . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12im-channel-ref . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12im-channel-set! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12im-channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-clip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-clip-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-close . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35im-close-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35im-collect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12im-complement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25im-compose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38im-compose-channels . . . . . . . . . . . . . . . . . . . . . . . . . . 38im-convolve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32im-convolve-channel . . . . . . . . . . . . . . . . . . . . . . . . . . 32im-copy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-copy-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-crack-edge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37im-crack-edge-channel . . . . . . . . . . . . . . . . . . . . . . . 37im-crop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-crop-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-crop-size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-delineate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35im-delineate-channel . . . . . . . . . . . . . . . . . . . . . . . . 35im-disc-dilate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34im-disc-dilate-channel . . . . . . . . . . . . . . . . . . . . . . 34im-disc-erode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33im-disc-erode-channel . . . . . . . . . . . . . . . . . . . . . . . 33im-display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38im-display-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . 38im-distance-map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35im-distance-map-channel . . . . . . . . . . . . . . . . . . . . . 35im-divide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-divide-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-fast-channel-offset . . . . . . . . . . . . . . . . . . . . . . 12im-fast-channel-ref . . . . . . . . . . . . . . . . . . . . . . . . . . 12im-fast-channel-set! . . . . . . . . . . . . . . . . . . . . . . . . 12
im-fast-ref . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-fast-set! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26im-fill-holes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35im-fill-holes-channel . . . . . . . . . . . . . . . . . . . . . . . 35im-flip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-flip-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-gaussian-blur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30im-gaussian-blur-channel . . . . . . . . . . . . . . . . . . . . 30im-gaussian-gradient . . . . . . . . . . . . . . . . . . . . . . . . 31im-gaussian-gradient-channel . . . . . . . . . . . . . . . 31im-gaussian-sharp . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31im-gaussian-sharp-channel . . . . . . . . . . . . . . . . . . . 31im-glcm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30im-glcp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30im-gray? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 19im-height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 19im-histogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39im-image? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-invert . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-invert-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36im-label-all . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36im-label-all-channel . . . . . . . . . . . . . . . . . . . . . . . . 36im-label-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36im-load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18im-local-maxima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-local-maxima-channel . . . . . . . . . . . . . . . . . . . . . 21im-local-minima . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-local-minima-channel . . . . . . . . . . . . . . . . . . . . . 21im-make . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10im-make-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10im-make-channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10im-map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-map-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-max . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-max-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-mdivide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-mdivide-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-median-filter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31im-median-filter-channel . . . . . . . . . . . . . . . . . . . . 31im-min . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-min-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-mtimes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-mtimes-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-n-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 19im-nl-means . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32im-nl-means-channel . . . . . . . . . . . . . . . . . . . . . . . . . . 32im-normalize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-normalize-channel . . . . . . . . . . . . . . . . . . . . . . . . 20im-open . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34im-open-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34im-or . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Procedure Index 52
im-or-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-padd . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-padd-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-padd-size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-particle-clean . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25im-particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25im-range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-range-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-reconstruct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36im-reduce . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-reduce-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-ref . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-resize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19im-resize-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19im-rgb->gray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-rgb? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 19im-rgba->gray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-rgba->rgb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21im-rotate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-rotate-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-save . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19im-scrap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25im-set! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11im-sharpen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31im-sharpen-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . 31im-show . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40im-size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 19im-subtract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-subtract-channel . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-texture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
im-threshold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22im-times . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-times-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23im-transpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20im-transpose-channel . . . . . . . . . . . . . . . . . . . . . . . . 20im-width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11, 19im-xor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24im-xor-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Kk-channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15k-display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16k-fast-offset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16k-fast-ref . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15k-fast-set! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15k-height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15k-make . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13k-make-circular-mask . . . . . . . . . . . . . . . . . . . . . . . . 14k-offset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16k-ref . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15k-set! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15k-size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15k-width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15kernel? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Rradian->degree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41re-export-public-interface . . . . . . . . . . . . . . . . . . 41
53
Variable Index
This is an alphabetical list of all the important variables and constants in Guile-CV.
%2pi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42%image-cache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41%image-cache-format . . . . . . . . . . . . . . . . . . . . . . . . . . 41%k-edge0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16%k-edge1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16%k-emboss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17%k-gaussian-blur0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17%k-gaussian-blur1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17%k-identity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16%k-laplacian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
%k-mean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
%k-prewitt-x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
%k-prewitt-y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
%k-sharpen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
%k-sobel-x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
%k-sobel-y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
%k-unsharp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
%pi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
%pi/2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
54
Type Index
This is an alphabetical list of all the important data types defined in the Guile-CV Pro-grammers Manual.
(Index is nonexistent)
